From: Subject: NYU ELJ ARTICLE (1995) Date: Sun, 15 Jan 2006 17:03:18 -0500 MIME-Version: 1.0 Content-Type: multipart/related; type="multipart/alternative"; boundary="----=_NextPart_000_0000_01C619F5.950D5A40" X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2527 This is a multi-part message in MIME format. ------=_NextPart_000_0000_01C619F5.950D5A40 Content-Type: image/jpeg Content-Transfer-Encoding: base64 Content-Location: http://www.law.nyu.edu/journals/envtllaw/issues/vol3/2/3nyuelj469image1.jpg /9j/4AAQSkZJRgABAgECWAJYAAD/7QtqUGhvdG9zaG9wIDMuMAA4QklNA+0AAAAAABACWAAAAAEA AQJYAAAAAQABOEJJTQQNAAAAAAAEAAAAeDhCSU0D8wAAAAAACAAAAAAAAAAAOEJJTQQKAAAAAAAB AAA4QklNJxAAAAAAAAoAAQAAAAAAAAACOEJJTQP0AAAAAAASADUAAAABAC0AAAAGAAAAAAABOEJJ TQP3AAAAAAAcAAD/////////////////////////////A+gAADhCSU0ECAAAAAAAEAAAAAEAAAJA AAACQAAAAAA4QklNBBQAAAAAAAQAAAABOEJJTQQMAAAAAApjAAAAAQAAAHAAAAAwAAABUAAAPwAA AApHABgAAf/Y/+AAEEpGSUYAAQIBAEgASAAA//4AJkZpbGUgd3JpdHRlbiBieSBBZG9iZSBQaG90 b3Nob3CoIDUuMP/uAA5BZG9iZQBkgAAAAAH/2wCEAAwICAgJCAwJCQwRCwoLERUPDAwPFRgTExUT ExgRDAwMDAwMEQwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwBDQsLDQ4NEA4OEBQODg4UFA4O Dg4UEQwMDAwMEREMDAwMDAwRDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDP/AABEIADAAcAMB IgACEQEDEQH/3QAEAAf/xAE/AAABBQEBAQEBAQAAAAAAAAADAAECBAUGBwgJCgsBAAEFAQEBAQEB AAAAAAAAAAEAAgMEBQYHCAkKCxAAAQQBAwIEAgUHBggFAwwzAQACEQMEIRIxBUFRYRMicYEyBhSR obFCIyQVUsFiMzRygtFDByWSU/Dh8WNzNRaisoMmRJNUZEXCo3Q2F9JV4mXys4TD03Xj80YnlKSF tJXE1OT0pbXF1eX1VmZ2hpamtsbW5vY3R1dnd4eXp7fH1+f3EQACAgECBAQDBAUGBwcGBTUBAAIR AyExEgRBUWFxIhMFMoGRFKGxQiPBUtHwMyRi4XKCkkNTFWNzNPElBhaisoMHJjXC0kSTVKMXZEVV NnRl4vKzhMPTdePzRpSkhbSVxNTk9KW1xdXl9VZmdoaWprbG1ub2JzdHV2d3h5ent8f/2gAMAwEA AhEDEQA/APSLftduY+qm8UsrrY6NgdJebO7j/wAGmLM1j2sfnMD7J2NNQBdA3O2D1PdtaiVf8pZP /FU/lvQerdPqyce61tJuyRS5lbBY6vf/AIRlLn1vq9vqtb+ckpMKeof9ymH/AK1/6lSLc1sbsmsS YE1kSfD+eWIOmQWf5FsAssdZaG5Z9hIqY6zb6rG2b2+702f6H/SWomJhvY0Yv7FdVj2hldr35DXl rQNm7+cfbvrbddXvZ+ktq/wiSnW3ZO7Z9qp3Tt27NZA37Y9b933KRrz+19f/AG0f/Sy5uvp+XdiY +Pl9AP6Brg3ZlhpaC37PHq+s62637PVRX6lln83/AMUiO6dY+xps6HkO9Y7bn/b5LAX6u/pA/Nts u/Q/8Wkp3/S6h/3Iq/7aP/pdRcM5rmtdl0hzzDWmogkgbob+n93tRcPDowqG4+OC2ppJaHOc86nc 732ue/6R/eWb1ajJuzmgYD8qj0tn2iu/0C3e4vurdtsZa9v6DGf7G/8AoxJTf9LqP/cmr/tk/wDp dReM6sbn5dLR4uqIH43rCb04mbD0HIZY8N3Mbmt2yBWPzckMb+f7mt/SekpDpcYLMI9DccW5/pZN L8n1bG1CrY3Irusu3b/0lmP7LfXSU7QOY5waMyguMgAVmdPpf4dT9LqP/cmr/tk/+l1gZGBbkV47 HfV500NdXXGW2vYHgV3ONlNnqWetXUz9L+kvf/hPS/SKX7Oe665z+i5Bdu9lhzi4PIcHscJyN9Vb nOf+b/1pJTuel1L/ALk1f9su/wDehR6Pl35nT2X37fVL7GOLAWtPp2WUtcGOdZt3Nr/fT9Krsqw2 UPxziir2V1mz1jtHDvWPud/b96H0H/kuv+vb/wCfbUlP/9D0Wyl92Xm1MsNTn0UtbY2ZaZyPcNpa l1Lp9+W2p7c23DNIJe6gw14OxxFrH7vZ+j/43/hfT9Wu1snHsybs2iuw0utoqa20bpb7r/c3030v 3f8AXE/VMK/Jqa5mS+g1MfPp7hJIbtsYG21fpa9n6L1H2VfpLPUYkp5k9Rua6P2j1P3sDnk4zRte 71z9F/0PQ9H9NR/w2J6f+FR8nOyK3WZdOX1ANsvrfbjGoDYy1p2sp+2D03V1/ZLHfofz7/8AhKdj M6hbY1t46nms3Frtr8QkAkWNex2z9F+icd9v+Dq9P/RKTMr7bkYP2TquYN9dYD/RcabXO9V9eTYd zWt9T1fobvR/m/X/AO03ppSsbNybyyr9rZ1dtr/SYH4tR2un02ufZRXZQ+t/0vUrf6f+lXUU03Mf usvdYNoGwhobIAG8bW+p7o/fXIYHVMhzmF/UOo7WhtjRZisO6ff6DvRdc/1XfQ936L3+n/PfzfUd MsORScxuTdfTke6plzG1msS72Bjaqbf+3t6Sm6qB6fku6azCOZZ6rA0faho923/SQ7dud+fsexXj 5LPd03Lf0uvCOY4XsLC7Kbva47XB7v8ADep72+z+fSUiHSeraT1i90cfoqRr4u2Vs/zFDJ6H1C9x LOr5NAcGhwrjXa2thPu3bNz6/U/Q+l/PW+p6n6P04Ho3XBWGM65a12vuNNTiRGjfd+67c9Gu6T1I ustx+p2ssfBDHAOrltbah7J/Rtse11tvo7P5xJSj0rq+n+WbuCD+ho10Lf8ARe33Her2FRkUUCvI yHZdkk+s5rWkgnRu2oNZ7Vms6R1v2mzrLyYO9oqZtM7fbptd7Nv0vprQ6fjZWNQa8rJdmPmRY5rW GIa3bDP5TXP/ALaSm0s/oQjpdXm6w/fZY5aCp9HYWdNoaedpP+cS7+KSn//R9HyKH5FmZRXa6h9l NbW3Mnc0k3e9kFvuao9Rw8i2tlzMuyg47S5wqGjyDXZ7mbv+Cezb/wAKodSrznuyK8ZlrDkUsYzK pNUscDZu9l9lP735qBnN6rlek6mvJxHVg7jWaCCTtO7acnb+Z6f6T1P0V1n+G9OxJTj09XY7p+Oc jqmSzLDzbbtxnSN9Y3s9FjfSyKKLLvV/7UfpP0f+DUz1LMod9nv6rkUW0ONNwZhVvrOwb33U+jXY 70Pc2mj/AA379Hpq39h665ljbrsq7dtFJ249ZqDXttftvoyK8l3q1MbQ/dd/wn/Bp2UfWwl+/Kub 75qIqxXSzbtAsHqs2v8AV9/sSUxwut0C/HFvVrr9v6O2t2G5gse9+yqbWUba3VfaKK7PSds/RepZ /hV0q58YnXXNr9bIzHvquFoewY1ctDG1nGexlvp2Uus9S79I17/+20H7B9aQ7czqOU0kgvHoYjgY dvc1ofe7027P0XsSU9KqD+nZjumMw251jcistP2yJedrt8OaXe7c39G/emw7upU47Ksqi7KubO+/ 9BXukkt/Rsv2t2t9irNr6w3p1GKRc/Kpcxz8twqHqbHb3i2mvLb/AD1f6Oz9Js/89pKZjpPWC4+r 1ex7XVvZtFLGe5zXMquDqy33U79/8tMemdfa5gZ1clgJ37qKydvLWt/Od+57/wDz4qNeB9Z6w1hz sm1k+8OZjNMH+cLbRc+79/Yz1PZ/g7avSRDh/WH1KrhlZJsY3ZYCzH2WNLrnn9C3IbXVZ+kx2MvY z1P1b/CeskpsDpHXWxHW7N0e4+hUQSR9La7c1vv9/wD1C2VjYTeuY92/JsvzKgwsFJZjV+4lpZab GXb97Gtez9x+9XvtmX/3Au/z6f8A0skpuKv0/wDoOP8A8W38iH9sy/8AuBd/nU/+l0XCZZXh0Msb se2toc2QYIGrdzfakp//2QA4QklNBAYAAAAAAAf//gAAAAEBAP/iAaBJQ0NfUFJPRklMRQABAQAA AZBBREJFAhAAAG1udHJHUkFZWFlaIAABB88ABAAWADMAMmFjc3BNU0ZUAAAAAG5vbmUAAAAAAAAA AAAAAAAAAAAAAAD21gABAAAAANMtQURCRQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAABWNwcnQAAADAAAAAJGRlc2MAAADkAAAAcXd0cHQAAAFYAAAAFGJrcHQA AAFsAAAAFGtUUkMAAAGAAAAADnRleHQAAAAAKGMpIDE5OTkgQWRvYmUgU3lzdGVtcyBJbmMuAGRl c2MAAAAAAAAAFkdyYXlzY2FsZSAtIEdhbW1hIDIuMgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA WFlaIAAAAAAAAPNRAAEAAAABFsxYWVogAAAAAAAAAAAAAAAAAAAAAGN1cnYAAAAAAAAAAQIzAAD/ /gAmRmlsZSB3cml0dGVuIGJ5IEFkb2JlIFBob3Rvc2hvcKggNS4w/+4ADkFkb2JlAGSAAAAAAP/b AEMAFBERGhIaKRgYKTMnICczJxwcHBwnIhcXFxcXIhEMDAwMDAwRDAwMDAwMDAwMDAwMDAwMDAwM DAwMDAwMDAwMDP/AAAsIALcBrwEBEQD/3QAEADb/xACSAAADAQEBAQEBAQEBAAAAAAABAAIDBAUG BwgJCgsQAAICAQMCAwQHBgMDBgIBNQEAAhEDIRIxBEFRIhNhcTKBkbFCoQXRwRTwUiNyM2LhgvFD NJKishXSUyRzwmMGg5Pi8qNEVGQlNUUWJnQ2VWWzhMPTdePzRpSkhbSVxNTk9KW1xdXl9VZmdoaW prbG1ub2/9oACAEBAAA/APs1VVVVVUWtraUWtra2thbVbC2tra2lVRa2lVVFpRaVVVVFhbW0qi0q qv8A/9D7N8vrjnyZseHBk9LdDLknLYM39qXSYsf8vL/7M5EfsfXf+lX/AKAx/wDwRf2Pr/8A0qj/ APK8P/rlf2Xr/wD0qj/8rR/+uU/sv4h/6Ux/+Vx/9eJHS9eP/RmH/wAr/wD2av7N1/8A6UQ/+V// ALNT+z9d/wB/D/5XP/4wT6HXf9/j/wDlc/8A4xT6HXf9/j/+V5f/AIxX0eu/77H/APK8/wD8Yp9H rv8Avsf/AMrz/wDxin0+t/73H/8AK8//AMar6fW/97i/+V5//jVfT6z/AL3F/wCaMn/41T6fW/8A e4v/AJXyf/jVfT63/vcX/wAr5P8A8ar6fW/97i/80ZP/AMar6fW/97i/80ZP/wAar6fW/wDe4v8A zRk//GrPp9d/3mH/AM0ZP/xivp9d/wB5h/8ANGT/APGK7Ov/AO8w/wDmnL/+MV2df/3mH/zTl/8A xijZ1/8A3mH/AM05P/xiuzr/APvMP/mnJ/8AjFHp9f8A95h/805v/wAYL6fX/wDeYf8AzTm//GC+ n+If95h/81Zf/wAYL6fX/wDeYf8AzTm//GC+n+If95h/81Zf/wAYL6f4h/3mH/zVl/8Axgvp/iH/ AHmH/wA1Zf8A8YL6f4h/3mH/AM1Zf/xgvp/iH/eYf/NWX/8AGC+n+If95h/81Zf/AMYL6fX/APeY f/NOb/8AGC7Ov/7zD/5pyf8A4xX0/wAQ/wC8w/8AmrL/APjBdnX/APeYf/NOT/8AGK+n1/8A3mH/ AM05v/xgvp/iH/eYf/NWX/8AGC+n1/8A3mH/AM05v/xgvp/iH/eYf/NWX/8AGC+n+If95h/81Zf/ AMYI2/iP8eH/AM05v/r1dv4j/Hh/805v/r1dv4j/AB4f/NWb/wCvV2fiP8eD/wA1Zv8A69Y6fqOq j1X7P1BxkHHLLE4ozh8M8eL08nq5c/8A3r6qq//R+zeCZ/8ADsY/8o5v/dn4a96oVKqqqqqqqqqq qqqqEqqqqqqqqqqqqqr5cv8A1aR/8RP/AN3YH1FV/9L7N8+f/H4//E5v/dv4c96VePrOsx9HDfOz ZjCEI+aeTLL+10/T4/8AvXPD1OWWSOPLiMNwM4yEhmh5fT/8Hy5MX9nN/N/9l/5X/g3U9Q7y6jHC BySkBAGpTlIRhGUT6OX1cn9r+9/J/wDHuf8A1p0f/f4v/NuP/wCCNDrenMPVGSHp/D6m+Hpbv+79 X1PR9VP7Zg9P1vUh6XHqbo+nuv0/+I3+l/dRLrunhETlkgIS+CZnDZPb/wB3k9T+a3+0YriN8bn8 HmH8z/2W/wC9/wDMLWPLDJeyQltJhLad22cf7mPJ/wB1l/8AKLydL+I4erkI4vtQ9WPwf2t8+m/t 48nrf3Mf/vN/w/Vfs/Vf+DtYOvxdRmydPA3PEY7uPNu/u+n/ADP5n7Pl/kdR/wCk/UOcfxTHLN6A jLcZ5cP2Nu7BDH1WT/ef2s2LLj/Z/wD5p9Bvquu/Z8kMfpznLJu2bPT+KEfVy4//AArP0/p/yf5y On/EcOaMpk+nsnLDMZTGG3Lj/u4vU9TNgy/y/wCb/IzvVDqMeSPqQkJRF3KJEoeX+5/MTlzY8Md2 SQgON0yIR/8ANmVE+ox45RxylGMpfDGUhGcv/FYv94k5YAGRkKj8Wvw/+N/7tHr4923fGwN+3cN2 z/vPT/7n/wArOoIIsagtKrz58+Pp8Zy5TUY6k/v/AL15cX4ick4QnhyY/UsQlk9P4own1Xp9R6XU Z8/S5fRx5P5ObA95lGIsnQMevj/ij/ygoz4zHeJDaPtWNn/mxRnxmHqCQ2V8e4bP/Ni+vj2iW6NS +GW4bZf+K/7xPqRBGo1+H/F/4v8A9BpjIG6N1z/U+bi/GMGb+1cjtnl2x2bpwxT/AGX+X6uTD/f9 LJm6f/yhj/8ACPQ/keucH4tjzyxxMJw9YSnglk2bcu2PrZf+Fz5/S/lfzf5/7O10v4pj6ucYQjOJ lCWUepHZ5ceWfQZceX+Z/ewZ/wDc/wDlXEnL1/pZv2cYpzntOTyentnCP8vJ/e6jB/vf9y69L12H qMMeoidsZAnz1D4D6WT/AM05f5L1RkJjdEgg9wxPJGFCRAs7Y2dv/mv/AL3KpywEhAyG4/DGxv8A /MeNoziBusUz6sLMdwuPxajy/wDjP+7dkK+Yf/VmD4YJD/0NjfUVX//T+zfOyf8Aqwx/+Jzf+7eg fRVXzfxLpZ9TDGcdb8WSGeEZfBk9Lf8A+D5f73petiy/3mMufq5wkcWHbIR8u+ePf60pQxepj9Ke fD+yYMXq5v5/UYM/Uel+z/szh0HQZeiySxTAy4pCOf1Ph29VH+Vk/k9Vk6jNly9R6ePrfX9f/wAH 6j1f++eP9h6mPR4cfpkZYZd89ksW+GKObP13/g2TLk9DL/cx/s+HP/v3rGHOJ4+oOOUtksm6OSWD 9py+rDB6X4h/4Lk6f8M9bB6WTof+I/8AVf8A/K7lh6Xqccx1GSANZc+T0Y+nDNsziGHpeoyejPB+ G5fxDD6eX9ozZ8/Uf+D9V/xDM+j6qchOEDjM8/r7R6M/2bH6M/w/9o6nFlyejl6rrcuT9s6jD0/7 R/vv/Cf2psdDlHQ+mMZHUYiZ4Nksf9/f6v7Z03q+jgw9Jny5cn7R0efB0/8A4P63Tfs//DvrdJiG HFGEQRp5t5/m+pL+7k6nJj/udVmy/wDEZv8Av0dRvxY5TwQ35QD6cdI+eX/eZMuTB/K9b+d1P8// AOaXyR+H5ujyYc+HflmLx54y9CEvRy/+E9T/ADf/AAf1c37fk/a/+I6n+f6v/hP/AH/vCMQbrx/5 38zI+T+JdNk6jPhqMzCBmcmTFOGGeP1IT6bF6OT9o6fP/Jyfzv8AxH/f/wBh4z0PUQ6f9lOLcY5c cxnxeh/OhGePLl6zJi67J/6sPSxej1Hr/tHr9R/6UdK7y6fqM2L9nEDCOSR9bLl9Dd6UYw/u9L+D dR0/q/tvp4vwz1cOf1/Q9b+x/wCDvPl6TqpYYY5YzPLg9XHhy3h2ZcVQ6bF+3dF1XUf8J+KdL/K6 n/0Z6b0f/K7cuhzDHmx5MAySzRG2UPR2Yv5OPFi6L/wnLgzYsPQdVi/8GzYP2j/0o/4lxyfhXVbM 0ok+tKMMMt8t2HrsX7Nh6XJ/vPWxdfg6rH1Ofp+sz+h/d/8ACP8AwbPn9Dc9HmHUyyejuxn15yhl OGe6c4fsuPL+G9b6n7bi/wCtcWLpsHU9H1GD9n6b1cv/AKbvt9PQxwG3Z5Y/y/8AutP+G/l/yv5H 9l6FV878V6WfWdNLFjrfcZx3fBuxThl9PL/5SzemzHq+okLj08gaMpRnkxbd/wD6T9NkxZeo9T1s v++zfs/T+h6vUf8AEfyM3B0fQZsWSWPqYjLhzgZM2nkxdZA+p/M6fN1Gb1cWf/yhg9D+V0n/AINg 6dyyfhkp9JOMcG3LLLI/D0++OCWeHX/+lH7PlxfsuP0PQzdR/fxfs/8Awv8APWPRZoyjKWIyxxyH JPD6fS4fWjPH6fTfy+l6j0cv/VeXH6v8/wDn5/2n/wAoNno8pO84rxesc37PtxQllxzw+lj6jLi/ aPQy5en6v/c5vR6jqPS/aP8AiXHN+H58spzx4tsJZsGWGKfpy9P0hP8Abuvy9Fi6j0P/AAj+Xh9D D1H7R1Ppf+E+g6Q/DJ/skoSxkdRjnPNgnj9PZ68p+r03/Vnq5PR6T8P/AJWL9o6LP6H/AMsvsdFg 9DENNspefL/HPPk/4nLl9H+T6vq/91/4P/6L9N/4N6C5sMYwlLHjBlU9m2MN+7L/AHfTyZf2fD/P y/zuo/n/AM98foeh6jFPpzHHOHpiUc88s4ZsfnEPVx9D037R12XpcufL/vsH/V//AKc/tD9BDHHH pEAUBDQf7uP9rH/4rA+L+I9Fm6vqgY74RGOeP18M4Q25Zyhlx+pj/aMOfN0v8v8A8Iwf/bDzz6LN PDgxfs9HDKW6WIdPkxbNs8fr9Lj67qMP/HZvSzfz8Hr/APox/wAThwO0+gz9V08ek2+nA7pZck4Y oepO/wDwb/wL8L6j+Vm9XL+2/wDcf+C4P2j+/wBR07h1XTdX1OPHPJhJzbPSnGRw5unzef8A8Jx9 Tj9TB+wf8P03W9P1vQ/z8Hq5um/9F+nds3Q5JQzYzg3ZsmSU8fU74fDI/wDgPU/tGXJ+2dL/ANVY fSwehgwf7n/wb1/Xzoyfhuacsk4RqOTLEZcM9ksWbBGeD/1J9PjxT9HF1X8vLn/n/wDEfzv/AAf9 o9BMekyRy5f5O+Eo5twyel/cy5fV9PoutxfzsvS9d/e/8Owf+o/0sP8A4T/uH6CIoAAU2h84/wDq yH/iD/7ug+kqv//U+zfNn/6sIf8Aicv/ALt6F9FKqr4vT/ihHqS6oxhDHlPTRnET27/+86nJ/Pxd Li/m4/73/g/T/wDpS9k/xDBiMhOVGJjjI2z/ALmUer03T9N/L/8ACs2fF/uOi9dzP4p08TXmsk49 vo59++MIdTkx/s37P63/AAuXFn/8R/4lR+KYCIGMjP1AZ4/Tx5cm+Mf5eT/hsGb0/Ryf38WX+f07 uerxwjCcpUMhjjx7hKG7Jk/tfyssPWxZf/HMz63FhmMUydx2DSE57PVn+z9N+0ZcOP0em9fL/Kwe v6Dlj/EsOQgR38yH9nPt/lep6uP1MnTf3v8AwbL/ACP7/wC0Yv2drB+J9P1EhHHI3Mbse+GTF6kY /wDpP+048H7V/wCYHL8T66fRenKIhU5elKWaRxRhKQ9XFl9T0/7P8rJ+0f8AmFzw/ioEDk6iWMR3 DFHJ08p58XqTHqfzcn7P/K/uY/539j+a7z/FOmgaMjYmcPlx5cn8+P8Ac6bH6XT5vVyf+JU/inTC ImJ7okHL/LjPN/Kj/KydRk/ZYZ/2bD6vqYf5/o/2s3/cZ3Q9fgGQQ3WZbNu0GcP5v/C+p1GLH6GL 9o/9FvXz/wDhP/ouxD8S6eeOOeMjsnIYoS2ZPNllL0/T9P0/V/ufyPW/seuuH8U6bMRGEjdSoGGS Hqel/wATj6b1cf8A4Vmwf+jGDp/X6hrp/wARwdQYxxyJ3A7PJkhDJs/u/s3U5cWHDm9H/wAoveqo V8zqOunh6gYjGOwwlllk3S3whi9P9p/8GxYM3q5f5mL9n/nu0fxDBLYIzB9Xd6R/7z0o+r1P/ivQ /wDKzlL8W6QWd9AAy3bcmyUIn0smXp8np+l1WL1cn97B6+BJ/E+mjCWQzqMCIz0nuxyn/a/aMWz1 sPrf7jNmdcPXYcwlOMtIf3Nwli9ONer/ADcfVY8GXHi9L+cnJ1mKGOOWUvLKtm0Gfqbh6uP0sWLH 62X+V/O/kuM/xbpMfxT7DJ8M5fypj1cfUfysf9n0v98mf4r0kJHGcgjIH0zuuEYz/wC7yZcuP0cb X4j1f7FglnIsR2/4fjlDF/c9PP8Ayv5v/cZ3HB+JgC+o2x3SjjxZMROfD1Esg/l/s2XFj/uerjy4 fR/8pusvxDp8cpY5S80DCE47Zynvyj1emx4seLH6vVZeow48mX+R6/8AaWf4l00BEmd7o+pHbGeS Xpf+lOTFhx582LpP/K+f/wAHY/626Uy2ievl+zPb/N2fs3qZPT9LHiz+ph/Z8yn8X6QSMfUG4ExI qe7fAwx5Onx4/T/ndV/Nx/8Ag+D/AMIUfjHSEGQyCueJ/Du9L+X/AC/5n87+T/7Mf+DukPxLpsk/ SjMb7MNusf5sP7vTeplh/e/8o/30Y/xLp8pjGErMzKMBtn5pY/8AiMf9v/c/7/8A7hOf8QwYJnHk lUgBIx2zl8cvSx+n6eP+b62b+S5x/GOjmQBkButvPm3H0sfp+T/vsfpf+PeyGaGScscTcoEbx/Dv Hq4/5n/inZXzf/Wl/wCYP/fz6aq//9X7N80/+rCP/iZ/+7elfSVVV8CH4XmG6OQ45RnnHUyFSlHZ Mel1XTfzf7n8r0/2f/zK6w/CpdOP5E/gyethGTdPbD0f2D/q/Ll9T1fR9L1P2bN/6L/yv+Ia6noc /WSxynIQ2ynKXpSyQnD1cf7Nj/Zet/3uXD6uTN637P0+D+10/wCz/wB/1+aP4XmiMMPIY4Y5Mf8A Lnl6b1fV9P8A8Iy/sn9nL/4P/wCE/wA/qP2jPlzt5/wqfVQGPNKhCAx4Y4vhjk3ep6mTL1WLqM/8 j9m6LD637R+0fyc//pf+zty6LqcuTFkkYepEQE8+OU8WXbCfq9d0+TpsWP0eu6TrfSx/yM/ofsGf Nm/4hnB0PUYp7qhuEsuW/Xzyh/Plny+nj/DfS/Y8f930PX/8d1H7N1HUtdN0GYDBjzGAhg2mIxb5 TyzhCfS4/UyZPQ/ZcXpZP7P/AIR+0Ov4j02fqJYjh2EY5erWXf5vLPDix/ysef8A7zL/ADf/ABTh +w9TkjPFIYoQyyj6scV/B/6PdRjyZenwer+Idd/Kw/8Apt6f7T/xLmPw3qjDF05MfTxZd0JxyZMW b9liM/S4seT9l6fB/wCGel1H9/Dn6f8AafT/APeh2w/h2ToMk5dEICOQQ8mUzh6U8UfSxZceTFiz 5uqx5vU9XqMGb0M/r/8Aoz/Pc8X4Vl6eeL0pRrHsj60TOGbJij/x3RdT0ezP0fVYc2XJ/wCDfzsH /V3/AJg/n6Yvw/NjhHp7iMWPIcwyCR9bJ/Mn12Hp8vT+h6OH+dkxftGb18/8jH/w/wDP/k4dN+G9 TgyQznbLIJZrxSy5JYdvUSn1P7Z0eT9m/wDBOr/m+h1H/g+bB1GD/uOpzOv4V0HUdDHHCW0iIMMk vWz5oy3fzP8Awb8Ny48HR9Ll9b0v53/cer/3z7qqqvk9R0/Unqo58YgYRhPFUpzhPdl2ZPU/l9Pn w4/5uLG8eL8DlCAgJCMZYpY80R/a/aJ4YdD/ANYdDi/kf3v5v7R/w/7Q7Z+l6zqOmOCUcUT6Zx7o zn8cxDF/L/8AA8P7Lh9L1f5X/hH+56f/AMruGT8J6nZlxwMZescU5ZMuTJ60ZYJY8v7J+0Yun/8A Cul/lf8Ag2f/AMG6jp/Vz/33c9F1GSJwyqOKcoyn/Oy9Xm2RH8zHiy/inT/zMWbNi6f+Rn/kfs/7 Z/6UfyMI/hfUQw/sw2SxwnKWCRyTw9RixS9T08mPquk6f/wXq8GbJ6v/AA/U9PnwZf2f+R6H8+z0 HVylvnUj6eLHuh1GfpZSyYvWy5c+XH0vSelm/mdT6X87/uvX/wDB/X9DDpm6Lqsoz4pDHtzmzPdO csX8vB0PqY+i/Z8Xq5v/AAb1sH/h/wDIz+j/AHvQ/n93UYsowjHgoyGyP834MuOOzH1ODqsnp5/+ I6X1P537O+RH8FywMp49sP52LqcfTQlOXTw9Lfi6r/wn0P5X7b6uT+x+H/8Ag/pYen/4f+x05+gz ZxKcowM5zhPb6uSPo+lj9Lpeo6H8RxdN62L8Q/av53rfsP7P+z/+D/8AlbOOm6Pq+kmctxyzyRxx zepOePz9PvxYsnTZcfT9R6mL0cv/AIR62DB/4R/4T/v/AEGB+H9VjyzyQEPMcezbmz4MUZYMeDpv 5n4Z0uDNgy4PV6f+zm6n/h//AAf9o/3zt0X4b6WSWbOAZ+rlzYtuTJOGOOf/ANN8voYMef8A8rfs 6IdD1OPpI4agZjKMv9ye3045/wDrj+7+x+r63q/+D/2P/Tj/ANN3nH4TnPUeoTER9f8Aar9TJk/l 16X7P/1bmwfsX7X/AOn3r/tHTf8Aot/YdIfh/Vxniy1jJxz6jJs9XJtl+2ep6f8AN/Yv9z+05P8A cf7r/wBOP5HVh6fqf2mXUZdgvH6QEJTl5oZM3U4v7nT/APddR6Pr/wDlL9p/Zv5/7P0+fTdBPH0B 6TPGMz/M0uXpT9Wc+pxfzPT9bF/Ny/3/AEP/AAf0v2h6vw/BLpcEceSt4+OQkcvqZZf3epyZMuPD m9XPm/m/+ZHuV86v/Uj/AOYP/fz6Sq//1vs3zT/6sY/+Il/7uwPpKhVVVSqqqEqqFSqqqqqqqqhV SqqhKFSqqr5l/wDqS/8AMH/v59NVf//X+zfM5/Evdg/6Wb/4m+m+ZggYdXmFyNwwz80jLzSn+I/2 v+6xfyv7OF9Jzy445YGEhYIMZD/DJ+U6TrOox9L+xbj6soRy4s3m8vS58f7d1PU+p6nr+r0P/hHR 9P8A2P8Awj9ienD1uXpenhDHt06T9svJvyT9ePp5MvqZP2j+zn9XJ/7L/wDsv/4OuXqM8J58mKW2 ZHRfFeSH8+c+my+ljy5P5X93/wBB5f8A0Y/8JdJ9dk6fNmEIR9U5OnwmUp5PSlLPD+V1H7N/P9H/ AHX8nB/8s9R6DoepyHJCWehPFlyY5nFv9OcP2TP1+PJ+zep/3Xpfycv7R/Px5v2bqf570fh/W9R1 MhLLjEMeSIy4pXD7Wz/wf0/2jPm6r+T6f/hvodN/7LdO8XXY5y/EBjhGWQSxerLEM+TpoepHJDpv 2j+Vk/7n+T6P/lRcvW5+gx4gBHbM+jKObIc2Tpuqy+p1XTdP1PU+p/wmH1cX87/iOn6b/wBKP5Dp m/EupwzmZRxnHhlDFm2mfq5PXGD0cnRYvT9HF/xOP+Rn/aP/AEn/AGn/ANGM9/8AWWWJhlMYnDly T6fFGO7198fWxdL1GXJk/kelny9Nl/keh/4N6uHqP2nP/PZ6X8Uy5TjB2bpZPSy49uTHl6fdg/bv Ty4uq/3uHPi6jB63/DdTgx/+Deh/P/Z56T8T6nqoYdkMd5RlluMpw2+hl/Zsvp4/Tz/7rJ/I/n/7 3/yh/POL8V6nNjnmhhJgYmeD+OUon+Xg/uZ/2rLn6T/w3+R/Y9P9m/n/AN99DoOqHVYd+4SNyjL0 xOPwn/eY+q/n4cvo+n/Jz/8Aodn/AK46bi5/+aOp/wDrZ3wdbi6kmOPdcfi348mH/wCisWB60JVk i9Hx/wAHnkmOo9SRmY58uMSyfFsh6P8A4nDi/wDE4MHT9O+0+V+MZ8mHDEYjtOTJjw+p/wBzHKf+ I/8AfP8A5leXMcfQdThiJZvN6kj/ADDnw5owxz/af2jH1/WetizdH/K6v1sOD/03/nf+i+Y/GOoH Tz6iWE7dsc8PMYQ2Tl6f7Nky5sf8zq/S9PP/ACMH7Pn/APd+8/xPPjEoSxg5RlGCOwzyYvNh/wCs /wBo/tftmX0Om/4jDhwLh6zq8mXEBjjCM8eXJkw5JH1fUxThi/uen/Kxfzem/Z//AE3zZ/8AuOnY wddnlihHpse+seLNOOTLL1ZeucnpY8XVZcf83N/L9bqc+f8A8H/9F3X8YEgMFSlG82LEfTyTx7sW U/zceT9lyYP+7/8Aef8A9F3k6vJPoM0scJzyY54c2aWI5J+p0/oQ/wCIxddm/aOt6XDn/sdP/wCn Ljk6vqCd+M1DH0sepxj1cm7dMT/m9T/L/wDD8v8AL/s9b/I/lYuo/kev1Dr034hnwCOKEDmMY48n UVPJnnuz/wBvH02Xqsf8r+Ti/bf5+f8AZ/5v7N03/pQ9M/xTNEZJenGoT/Z4R3z9XLnkel/Zv5X7 P/Kw+j1PrdR/xGf/ANF2h+IdV/LiMQE5zyYryynhh/LhPqcfWY8X7P1Gf0eoxY/7Gf8A8I6f0v2f /wBOGcX4pm6nHA9PjEpmEc84Tns8spz6X9n6XJ/3vq9P1OX18/8AI6f+T/3/AP4OMP4tlzbp48Jy Yqy+lKG/z+h6np+rky4//R/Lj9HpvQ/aM/Tf+jP9/wDkdv4b1Z6vEcktpIlKH8vf9j/0oxdVjwZu kzf+Uc3/ALMf799BKqhL5v8A60v/ADB/7+fSVX//0Ps3zf8A1on/AMRH/wB3ZH0nhHQ4fUll2+ed xmTKfmjL/d5Mfqejlxf+Uf8AcPTjxRwwEICoxAhEf4Yj08eNGXHHLEwlqJAxl9nyy/ufzMbz/wDV /T7PT2abDh77/Qkf+D/ad/r+j/5S9dyyfhPS5IxxyhcYCUYR3T+Gf93H/c/m4f8Ayj/6L/8Aoso/ C+knGQEbhkjCEtsp7JQxDH+w+n/M/wBx6eP9nzYHm6v8Nx5bj023eJ4cmePqZIS2YP8AhcWP0fXz dBl9L/hvQ/Z3ox9L0m6AjpKE5yhWWe/9ol/M6n1v/CPV6vqvR/v/ALR6/wD4P6v/AKLtdP0XS9Fk vDGMJzuMfMfh/wCKy4umx5cn8rF/vvQ6f+Q1k/C8GXL68hL1K2745MuLyf8Adf8Ag3UYfTxf+UnM /hPSmJjKBoiUJ3PJ8OWcOq6jJkyer/ezZceP9oz/APEfysXT/tH7NgePp+nwjqJ5MkoS3Sxy6eIz zyf24Q6Xpv2rp83oYM3V/wAvHn6fNm/aeo/aMv8AxP8AJ/aHtPSdJHJOUgBIiWXJHcdn80fsvU9b +x+p6OLJmw+ph6jrfR/77/wj++yPwrpdgEYkgmOQS9TJv3Qh6XS+n1vr+v6WHD/J6fF6/wCz/wAx 0w/hnT9PKBxxI9MSjj8+TZGOU+p1H8nJk9H+dl/98/8AcYGR+D9ILAx6SvQSn5Nxhl/8G/mf+Cfz ceP/AIL0P7b04Omx9LEwxCrJlLWU5SnP+Zky5c2b1s2XL/5VzvUqoSqHzY/hfTxxzxASMMh35BLJ l804y9X9o9TJ1Hq483q+n+0Z8X9/0n03HNgh1EDjygSifiiXlx/huGGQZKlKUdwgcmTJn9Pf/Ly+ l+15c/p+vic4/hPSiBxiB2yAhRnk8sN3q/s/TZPU9XpsXq/zuoxYPQ9f/wBGWp/hfTyvdGRuUct+ pl3+riHpdN1GPL+0etizYcX8n+T/AOUW/wDq/D5doIMN4iYzyQn/ADz6vU+plxZfWy+tl/nfzv8A f/8AhDmPwnpaj5D5R6UfNk/tbvW/Y+p/mf8AhXSer/uOo9fp3bquhxdVt9Xcdh3w2zyY/P8A7vL/ AOC5MH83D/uM3+4Yj+G9OIzG0n1RtySlPJPLkht9L0snU5cn7Z6Xp/7n12P+qOm8JUYjDXq5/wCx H+3039/+1/8ABMqT+F9MdpMSaAx+aeSW7FGXq4+m6n1cn/hWH1f9z1Hr/wDpP/wzR/DenIyQMSY5 T6mUGeTbKf8Ad9XH/M/k5f7f9j0f7WH/ALjAv/V2AGJqVxkZxl6mXf6kh6WTJly/tHrdVl9HH6P/ AIR6/wD4N/4N/wAOxH8J6URjEQoRBxx8+T+1KXrZOj6j+Z/4V0nq/wC4z+vgUfhXTXI7Pj37huns /n/yup/Z8fqej0uXPi/k+rg/Z8/oPV0/TQ6cEQvzHfIzlPLOUqhi/mZeqyZ839nFiw/+YnoVVSr5 gN/iJ/8AED/3dk/+Bvpqr//R+zfNj/6sJf8AiIf+7c76SqqvlfifUGMY4MdnJmPpxjCUIZPS/u9b 1OLJmn/K9HpMWX0M3/pTlwPl4Oun0nT5ul+DLgMdm/8AnSx9L1E8fp9Rl9LJn9b/AKqw9T/4R/6L /wAvA5YurGDPnMcwMZ5OnxT6r+XLbCWHPlydR/4Lj/Y/7uP9i6fPn6f9nwep/wCE+v6C/h2QeqBG YIl1XUxE9uOU8n8j+V1nT5dno4sv9v8A4fB/4T6v/vP1A6HPkyjpAMu/JP8AabySjjzZsGX0/V/l 5cv8/F6X8rL6OX+/6v8A4R/4N+zdO6D8S6uOI5DKjAw6TNv9PZj6v1J/t34h+1Y+n/lYcPSfs2Hp /wCRn6f/AMJ/4fqfQ/8ACPX/AAvNky45epOOQxnKG/Gd/l8mXHjydT+z9Fgy5sPq+jnzdP03of8A vT+0PgYYYAMsyem9MZ8m71ID1v2eGSHqfsvUep/6Sf8ADYMHT/8Ac/szpmwQyftE55BZ6jFhlIxw f2t/Qf8Ag+XJm6f1vSwf9z/w3/gH/hPTf8T6/Tg63NjyzjCWMwxnLjh0cZQh1EvQj/4P+zdFi6P1 vWzZen9b1v2j9m/Z8v8AwuD0HOXW5cmCOUdSKlLpp5Ti9P8A8F9Wf/hWLLk9P+V0n/s9/wCEf+CZ un6n9o6b9oe/perzZOrnjySuB9Q4fT2T6f08XoYv+Jx/+F4urwepkw/iGHP6+D9oy/8Ag/7P/wCD ev7aqqqqqqEqqqqqqEoShKqqq+bH/wBWMv8AxEf/AHdnfSVX/9L7N86P/qwl/wCJx/8Au3q30XCG fHOcscZRMo/FCMhKcf8AxuL/AHTury/teH1PS9SPqDT090fU/wDNW/1Wo+nnrLHbOr2TjU//ACnk 9LL/AObGvRhu37RuP2q837/22R0+IROMRjtPxR2jZ/5kxsYsfT5QJ4xCUQTslER27oy/melkx/8A lXH/AOgmvQxQqW2I23K6+CUv5mTJ/wCUl3YsmMzNHGbMr+CWP/e5MnqfycuH/wAq/wDcNYo44wEc YAx15dtbNv8A5T9L+T6TzRwdHMjZDGZfFHbHHu2wl6XqY/8AxOb+T/491PQ9PKzLFA3rLyQ8/wAf 93yfzP7n/oVuODFGfqCEROtm/aN+z/u/U/u+l/5RYPR4Nph6cdkjunDbDZkn/c9XLi2fzcv8t0h0 +LHOWSMYiUvjlGI35P8AxuT/AHruqqqqrnGcZfCQezoxKcYfEQLWMxMXEgj2IhOMxuibB7htUsGc Y6EgLGQkLibaSqoYjITAlE2C6KhUqr5sf/VlL/xGP/3d1b6Sq//T+zfOh/6sJ/8AicX/ALt659F+ b6fL0P7VGeKUIDCMmONTG/NKvV6rP1P++y9J0WLp+pzftnX/APEdTlzdT/6T9R1PvwyRyRE4G4kC UT/hl/byOXWbz0+T0v7myfp1/wB5sn6Xp/8AmV8zoj0R6XBE7avFsj/vf2qJx/7vF/O/bPW/4n/z N+0/+CvB0PVT6XBE45CQOPrMsMR2elHJgzf+Dfs3pY8HWZcOb1M38n9oz/8AlD+w3Hq+oljNdTAn f0/nx+n1Ev58/wBl6nFkyfsfQ9Hi/uftnTYfQz9R/wCDZnXo+szyzCM81gdRm6TZKOL+ZjxY8/Vd Pny+ljw5vW9X0sP8n0On/s/yP/Sjk6T8Qn6OOByx6ePp5ckcsceGGLLljnz4v2f0/T9D+Ri/nej0 P/hHUftX7S9sevlPqfRlITxT/aIShMY4Qj6H+7x4v+N/9KcPU/tv/Efzup6bpun6XC+j+H7f2DDu rb6WPdu+D+3j9T1H5rBLLj6Q9CL2SxDrPU0/l9LLF634l0/+4/vfiP8A4J/v/wCR12TqP9w9fS5R GOOOOccVdH6nqiGPfinGeD1cmTLkx5vUw5s3/EYP+/8AW/8ARr+wB12bLixZcuaWOXrxwZ4fy8Po 7MeT1PWyZsGH1f2jLj/bf53/AIN/4T+zfs3/AIO7dN+IZZZhinmAI6jPgEJDHulihjnk6L+XswZ/ 5HV+lg/k+h+0er+zuGDr+sPT4cs8oP7RIw3y9PBiwekM/wDL/av2frcPr9blw/7/AKHP/bzdP0/o f8Q9OPr+o3YTPIJRlsxzl0+zJilmlmz9L/4Rj6rH0/V/snXYsf8A4N1vQ/8Ag/T9Rhzf+D9T03oP tZ8s8VbMcsl/92cfk/8AGftXUdP/AOgXP9szf+k2X/ldL/8AjF6IEyiCQQaHllX/AJryel6+H/zS 6pVXDPGM8cozNRIIlK9nkr+Zk9X/AHf/AI18b8NydLDJlzYzDHDIMezHcYeTF/KxfiGTpvU/8F/b c3U48PTYf/B/7eH/ANGc7774f/lzQjL8PyEgaHHKP9XqYMX8v/zFlyYv/MjhmwYsHX4Y9EIwy1k9 aOKP8r0Nn8rqPxHpelydP/6N5MX7P/Y/aP8A5Xcuk6vLGHTyiRDDIY98cOPHKPqZ8ufH6fU9Fj/n 9Lh63+z0vWYP/Rn1v2n11xfifU+jPqcmbFtlj9SEI/zcmPLv9LFj/ZsWPp/5X8/F0fUev1HUf+H+ l/4R+zfyGcn4j1Xpy89enkzwyaYP2z0unx4+qxf+D+p/1b1X7Fl6j/1I/sXoZ/7X7P6H8/O9GL8U zZuouM4RxDJ6HpZJfzMkZbPS6jF0/wCzftf7V1H97p//AAj9n/Z//Rf+Rn6h1/HMUckunqOMzOTZ H1474bZY8/qfyv7uTF6vp/yf+/8ARc+lw9R0+aHTxlDFuE+ozww4/JL0s3T9Nj/ZvVyf+D+v+Hen g/4f/d5uo/8ACOpz+v0/Nj/Fepjgl1M545iUBkhijLfPBOeXHj9TJ6PT4PS6TocXUf8AhXr5+oz/ ALRi/wCKwftD0Hrs8cmXH6u6GHJ0n8yse7Ji6mXpdbj6nLix+jixf+VsGDBnw+l/5ncR+JdTlvFG cdc3UYoZpbMezF04x5um6f1P2fqMPrZ/U/33T/8AhHT4s3/swx1P4j1kBLJHLCoYsef+XATx5Zep PoMuP1M38/J0mfLjyZvXw/8Ae4P2b/0o6ntPVZc2fJH1YY9k/R9DLAT9TH6fq4+s+Pp+ry/tGb/h /wCfg6b9mx5f77yx/EcwjAQljxRhhwZduQjDiy+qP/CPTxfs/UZcnS4PT9Dp/wBh/Z/2fqcv/oz/ AMO+j0vVZpdVPFlOh9Q49ohLD6eLJ+zf+D9Zj/n4eqwep6H4j0fWftH/AIT/AMP1HT/8O+whKqr5 sf8A1ZT/APEY/wD3d1b6Sq//1Ps3zof+rCf/AInF/wC7eufRco4oR1EQPcGwABQ0AacB0+LecohH eRtlk2jft/7v1f7zy5snTdHtlkAifMYbMZnOMY/8Rkx4+lx582LFhxZP/CM39j+Z/wCV3XHhwZcY ljhE45Vlj5RtlKf/AKM+n6f93Nh/3zf7Liu9kbvd8I+L/wCCqOlwgRAhGonfHyR8k/8AvMf/AHWb /wArI/ZcVmRhGySZHaPtR9HJ6n/mH+V/4hGP0cglhgBtiTjyQ27YeaMMuTH6fp+jk/ldRj/82OfU jpukwH1YxjgjVw2fyvNOH/otih/3vptH8P6U/wC6x/w/24fD/d9P4P8Avf5zydT1P4f0s5QziEZy MckxLH/d1n6XUf2//CvRy+r/ADv/AEXTLqvw8w/a5HHtmfTOWUfNKf8Aw3p5P5frf2fU/vf+i/8A 6bO2E9HnE+lxxgYwIGXDsj6cZy/mfzMXp+h6nq4//N+J6P2bEJRnsjuiNsDtG/HH/wApZP8AdY3o ShUqrlOEckTGYBifLKJ/xf8AeY3nj0HTQiYwxQEZVvjHHDz7D6mL1cez+b6OV7HDNgxdQNuWEZi9 wGSInHd/4vKzDpMGKJhjxwjGXxwjCEIZP93/ADcePH/MZj0HTQIlHFAGPwSEIfy/jyfyv5f8n+bl yf8AmxodJgG+scPP/d8kP5v/ALM/99/vP73/AHrn/wBXdJt2+jjq72+nD4q9L+36f/du37Pi9T1d kfU/7zaPV/8Aln+8jL0uHqCJZccZkcHJGM9v/i/Vxpl0+I5BmMInIPhybf5m3z/y8eX+7/vMn/m1 mPS4Y79uOI9S/V8o/nbt/qftXk/nf3Mn97/vHh6f8Jhgy5pSEDiy7P8Awf0tmPH6X9r/AHmfDl/7 3P8AyP8AiP8Awn+Q9v7D0+z0vShs+LZ6cNm7/vPT9P0fVZl0HTzNzxQJNR80IfDHZ6eL4P7OH+X/ AOamz0eAmzjhdbPghu9L+3+zf2/7Hpf7hA6Lpxt/lw8muPyQ/la+r/4N5P5X83+d/JbxdPixSM8c IxlL4pRiISn/ALz+bkx/3XoVVQr50P8A1Yz/APE4v/d3XPpKr//V+zeCAvrsh/8AKOH/AN2fiT3q qq+D+LmMJxnGcsWaMZelm278GXd/c6DqfT9f1f7ePL6Pof8Ac/s3r9R/IeUR6icsMSJ4v2iEMZx4 55McOkn088HU9R+zYceX0eg/aPwrDl/ZsPo+vgz4v2fqP7/Us9LPJDPGW/JInP1OMQyTyejLDHHP 9ix+nl/kel62Xpv2fP8A+Vf/AJX5jlznp8s/WyHJ6cfVxxx58fo9VvwZPT/aerydR6XV/wA3Lg9D oMHT9N+z/wDhP7P0/odO7fiOU9Nkh08M84g445MeXJln5JSz+p1HU9dk/wDR/Fn6T+R02LN+0fs/ pf8AD4fX/aG+s6iQjllHLLbHNKcfNkw+ri9DH/4H03XdLj/8G/nftP8A1b/6L9T1OL/0Y/3+HUTl ij1WSBy48sv2eYs5N/pZR0nq+r/ufV9X9pw/yP8Awj/wbP03Tf8Ag2DqH0emyifWGOSeX1o5Mh9L z/s37LU8fRZ8mP8A4P0PS9LN63/Ef9YZGPxPqIQ63HeU4tuPJunGIybfVOD0seT1sfUYcXr+l/6C /wDTh4xmx5s3SmWSUo7uojPP5+mhmlt9XF1GL0cmD0sXrdT+xet/I/aPSzdN/wAK4wyxx9LjIzTx 4zLqPVy/z8m3Lv8A/Ubjy/sufBn6X1+k6nJ1v/pz1HpdT1XT9Q+hHNOXURw9RkyRlWD9m2RyY/2n b/6sv2npsP8AJxZM+X/icPXf+q/p/wDuP57wY88sPTwjGc9wn1EcuH1MuHNky+pDFi/Yep/sZfxH B6uLN/1Zm/aMHU+rn/8ABv2l+zVUqqoVKqhKqqqhKFShUqhXhxj/AMOyf+Jwf+7PxJ71V//W+zeH H/x2X/xWD/3Z+JPa8eLrcOfLLDCdzhzEfw/2/Ux/7rN6WX+R1Hof8P1H/g/Uva45ckcMDOZ0A15/ 914nkP4n00cccxn/AC5EwjOp7JSj/u8eTZ/5Sy/+P9J6seaGUyjHmB2S/wBUMfU/+6cuJ5hDCOrJ Micxhu9OUjsx4N0MOXJ03T/2cXr5sWL9o/8AFvVmzQwQOTIajHWReXD+J9NnmMePIJSN7RG/PtHq /wAvJ/ZyfykZ/wANx5spymUxuAhkjjnshmjD1PS9X0/53+9yf2M+B7oxEBtiKAaVKoSqqqEqySAL OgDh0/V4eqBOGcZ18Ww7nocs2fH08DPLIRiPtSZwdTi6mPqYpCUeN0T+/wDMehwxZoZr2EHaZY5/ 4ZwPp5MeT/yqsc+OU5Y4yBnGt0ftR3/2vU/7v1mD1mAZf2czj6v/AHV+f4fV/t/+KelKEvHDrMGT aITjLeZRx7Tu9T0h6uX0/S/7jFjdZ58eOcccpATnu2R+1P0/5mX0/wDxON3QlVQlVeHH/wAbl/8A FYP/AHZ+Kvcqv//X+zeHH/xuX/xWD/3Z+Kva+AMt9XDNg3SxQhljlhLFKH7NCPofyumx/s/T5/2v Pl6bF/4Fm9fP+zer+zeh0/oPtY8gyREo3RAlqDGXm/8AKWX+djXPkjixynI0ADb8xGBxdF08pZZk CWHfilCG3H6U4Zeq9THi6P8AbcWLovSyfzs2f/6Id+q6z0fU3TmYftMYy1nCfoTwer6XTdTi9D0s ODq8WTN/Jz9P/wCD9Ll/8I/8I/n8kyZAE5jkyQ6XLWfFOfm6j1IZcfT+ri9D1f8AufQz/wDdf+Fd M/RdRmieklkJsHGZX/VD+X/K/wDK3/cvi9D1GOEMMZ55yrH6c+nMdksXkh6v/hOHH0ObpcX4Zj/a MvrdR1GbP6H/AA//AIR/xHJPqMmHps15MonEY5YJznLz9FlzQ/8ACP8A4Y+n1Xo9R/w/UdP0/wCx 9N0//ox63Z6k4zlWSZ6I5ccfWGSc/L6ef9p9L8S9TqM+L8L/AOsMfR4eo639oweh/P6b1/8AiHGf U58YhklKZiMkxghGWSGbq+m9XB+w9R0+X+fh6/qv/TLrv/Vh+H/+E/8Apy3h6sx6j0sk8gMM/Ubt xyfy+lljn+xZM/qfyvR/tZ+mzZ/7H83qf+H9dxxZssOmxZp5p+nkJ/acuQ5MnoSxepi6fpseLoMn Qdb0uHPl/wCJzYc//EYsH7T/AODdR/4R7vTb59HRlOcts4jKYy6fNk+PFi6j0/7+HLmxel+z5v7/ AP6MPiYM49HBDJOePpzgqGWEskJft0Zel1OD1PU9bJl6fFjy/wDV/R/8Nn/sdN0/UfyMD9H0M8k8 GOWbTIYx335Zept/mepj/wB3l/8AKT2KqqqvP1MsccU5Za2CMjk3DfHZX831cf8AvP5X+6fG6LND Fnz5zkGTHM4f58Yn45mfTdN+F4/R9fF1XoYsnTfz8H8/B+0/+FP0L4/4yMJxwGaRx+ePpZ9P5GeM c+Xps/Uer/uf91/5k/3H/EPldP1Jnt9eXpYpTzxy9RgkcGDq8+KGHF03Wftv8j9m9f08/wDYz/s/ X9T0v/mBEOsyzjj/AGvLlxyMI+h6cZfzsvr5sX8zpcWP/wAL6vP0n/Vv/gef1+n/AGfqs3Ufs/8A Ocuo62ZxGs0pSGTqoRl6gwRl6Ef/AAL1Op6TH6ubN/bzdN+G4P2fp/xDPl/8pPr/AIZ1UeozykZA zlh6WcuPi/8ACsuX+V/u/wDicX/m7C8fXdRDH+IZD6ghKPSyju3DdHL6n7Tjx/zf996P8/0f7/7O 8+Pq8kcMhHPKUpYunzZJE+vPpt8vS/FsmPFh/nYsWDpfT/acOH/wjp/V/aP2h3j1OyOKGTPKfTyn l3dRH1MMfJGGXpvw7F+I/tGfrMuH1v2n+f8AtGfP1Hpfs37R/I6h5o9bkHSwy5ssxLJ0/UREt0/P 1GLJDL0Pp+l/6N+l6n8//uG8vX5I9RhMZyGvS7/UlKEMmLPGePqcmHofT/8ACsX8zF+09b1//hHT dTiw9P0zv0/VxOXFCOSoTz9ZGYjKMfU3evk6b+bi/n/3MnTfs/o5/wDe4f8A03eLpeqyYMXTw6eU iTDqd+KVy39Vixer02P0sv8AO/4vJ/Zwf+D/ALR/6c/tDrDqJ+hcepMomWLfvjlwQjL/ANGOj6n8 W/aOpzfh+Xrv/Rj0P/B+n6jF/wCi3Tdf/P8Aofw7Kc3TYp1IEwj/AHfNk+H+7lyf731v73rfyfXe 9CUJVXhx/wDG5f8AxWD/AN2fir3Kr//Q+zeHH/xuX/xWD/3Z+Kvcqqr434kcU5wjWQ5YCWaH7MYe tHF/Lw9R/wATkw4cuLP/AGPR/n58/wD6LvT0EsewDDGQgYxzxyT+3+0b8uX1OoyZM2bL1f8Avuo9 b/vcL3vKeqiOo/ZqO4wOUS+xtjOHTZMf/jv7f/m1nrejj12I4ZylEEg/y63+U+rj/mZMef0/52LH /wCa3A/icBP0xGcojJHppZoiHpR6iWz+Xl/mevi/m5fR9b9n/Z/Xyfs73ZcoxQlkIJEQZbY/H5R6 n8vH/wB6jp80eoxwzR4nGM43/DMerjaOSImMdjeQZ7f8ETDHkyf+hMbolVVUKlVVVQqVVXKesSAS D/FGv/fv8p5+h6KPQ4RggTKMb2763eY+p/useB7VVVQqVVVeHH/xuX/xWD/3Z+Kvcqv/0fs3hx/8 bl/8Vg/92fir3KqofE/FOmOaYrHMyiP5PUdPKHrYc8jP1ceTHlydP/Kz4vS/9N/5ub/h/wDiHzus 6LqeoxZYzxmWbJj6bfLybJ9Tgn/4T6f8z+Vl/ZOpx/2PQ6f+V1v/AL0aZ8MsF4seEbZZ/Uh08owl jzYv2b0ep/8AAv2j+z0/VY/X/wDH5Ojc4dJPJhGLpBKo4MuAzmBjl6/rQydd0f8AMyfysuf9l6rD /wCk/Terh/8ACX1PwnpvQ3mMJwjIiW3L6UPPH+VkyY+h/C4Yej6XH6WPp/5397qP/Q3Uc0+nyjqh lwQyYshyD19uz9k6jp4T9L9o6n+Z/d/ZP/en9oy5/wDwf/0Yc8XRZRmhCOExhi/a4HJ/L9PZ1M8m XofS/mevlw+ll/7j+R6v/j3nH4cTgAhhnjlHFHB1G2MP5+eOTpMvqY+my5PR6/Fg9Lqc/UZ//Wj0 +X9l/af+41wdJkxzxz9Ak+lnhH0xHp9mSWTJ1XTZP7nrfheXremyZP7Of/wDP1Wbp+n/AGf+f6HH i/D8k8GSM8EqJ6XJjxRGyO/Fvw9d6WP9o/8AST1MPrdR1H7T1H8rqep/v9O9I6PLikScJn00c+af 7PGAluxZccOm6HqOm6bLk9HJhwZv2j+T/uPW/aP2b/uKj0OSPUQkcWQGMsP7OITx+jg6fFDBhy9P 1P4jlx5+sxej/wCE/tPRYc/ofiHq/wDg3rf8Q/VKqqqqqoSqoSqqqFVVVKqqvDj/AONy/wDisH/u z8Ve5Vf/0vs3ix/8Zl/8V0//ALs/FHteL9sxnqP2UG8m31Zgf7uFwxY/V/8AKub1f7P/AHH/AJg9 ftcp5I44mcyBEAylI/wx/uPBD8TxzEchhOGGRqGecY+jPefT6f8A3n7Z02LP/wCi+bren6fB/Nxf 99070Zusx4uLl54YZ7P91PLs9P8Aaf8AusX/AIRi/wDNr1E1y4dR0mHqgI5oCYGsdw/f03m6nq8f 4cMeOOORjM+njGCMNm//AHfT+n6mD0v95m/8xZnXpur/AGncDCeMxqxljs+L/usuLJnw5f8AzDne xbVUWHj6zro9GIynGUt0hjj6YE/PP+1j+P8A3yek63H1QOyxKB2ZMcxsyY5/93l/+DYf5D1k1q8W L8Rx5ul/a4g7AJz21/M/lGeLJ/K/8xPRgzRz44ZY8TjGcb/hmPVxOypVXixdXizzyY4anGds/wCH fIf2/V/8o/8Aox/3Gd7Xh67rIdDj9bKJGN7T6cd+3d/b9T/yl6v8n/x+XC6Tz+nGJ2yO4wjUY+eH q/y/U6j/ALrFg/8ARjM54+shmzTwwEiYaSyUfR3/AMv1em/aP7P7Xh/l/wAj/wAq/wDj3tcoZI5R ugQRZFx/wn0sn/oXHlwucOqhPNLAL3REZ8eTz7/T9PJ/vP7b0vjn8bxDd/Ly+QmGSQxSnHFKP9z1 cmL/ALnF/O/9l32XE5IiYhY3EExj/Ft2er6f/ifVxf8Amx1VUqhKoSh4sf8AxuX/AMVg/wDdn4q9 yq//0/s3hx/8bl/8Vg/92fir3Pz2DpOq6XqRkJGSIx5d0hD055s+WePN/ves/wCKz+l/f/4bp+nx fsv7P0/+49vHIyiDIGJI+GXxR/8AKeT0v5Th+IdOeq6fJhiaM4yju/xV/L/8xPmdVHN+IdN+w+lL HKXpxyzls/Z8PpnBmy5MWTFk/wDC/wC16PT4en/96f2djP0xvPj9AyGTNhyGeOOPZlwR/ZcvUep/ Mw5sv/D9b+0f+nHVf+nDyS6HKIkTxzOCGfPIYcUccv5WXZ/1b1HTdF1+PP0eXpMHq9V/J/kfs/7T +0/7h6uj/DwOrgZ4jWPFCOPJk/mbMmLJk9L1eox/yMnV/sv7L/Z/kdP/ADum6Xqf5H87p/GYHJ6M Y45zEcnqy9H4owxQni/u4cmDNizer1P/AIP6H/dZnkngPUY548eHLEyyY8mLJ1B9SeLPUMWXrP5u TqM37L0OHosX9/Pn/aM/Vfs3/g/T/wBnAYMW/AMvTEj0MmP9njDf6c45MGP9o9PLl/lev6mX/wAN /wCI/wDCcXUdV1H/AKTg/h+aIljzYzkzSx4R02eH/otlwQ/m48nVep/4B6HWf+Get/60P53/ABHU /wDg6cv4fkyS6qIx3ujmqWTGIT3zyYeq6bH03WYZ/wDqQw9bmxet02LPh/aPw708PTev/wB/Ofp5 dUck59PkGLJm6fL6ewb9kYen+JZMuLpcvrf2fU/96M3/AKVf2N8nQyj1oIx5TtOL9nlj9P8AZseD BDDhydN1HU5cefrekxf8V+0dHg/9WP8A6Tvf+ORnPFCMIzlWXHOXojzwhiPq5MmP/wAq/wDpP/6c PH1X4T04y4Y+nLIJTnLPklvyfFD0sf7b1Pqf+lf7N/J/4f8AlZv7H8/14HS5P2j+1IZv2j1x1X2P 2Pd6X7P+3f8Asp/4F/1X/wDM7zdL0OTFiAGGUMuOPUevPb/xPqxz/sPR4vS/9WmX1f2bN/6T9P8A sv8A7LrHo8xOP1YZZE4+nj0/pRxx/Z59PD/wnp+t6nqunzZ/wv8A8M/nevg/4jp8v/ox6P7OuD8P h6w9TBLac/USl/Lntl0sof8AgXqf+UfW9L9nw/8AlLN/70HD02YnpZSxS3Qjixyjlh5P5Wef9vqM X878Ky9PhxftHrf8P+I9P6PT/s//AKLZ/sFefqIznilHEdsyDsn/AAzr+XkfC/DcHUfh+bKZ4wcZ OHHEdPGW3dsx4v2jH+1dR/w2D/1oZ837Rn/aPW6j/v36V8X8eEp9JKGOMpynKArHHfLyTx9Vk9T/ ALrF6fTf+6Xn6vqp9VsrBmEYZMM90oShLdDJj/l/s+L+9i/Zf2nN6/8AY6fPiwf9/wDyPP8A2WYG UnFkED1E8mWGPF/MzdPIZMf4b6WLrsfo9f0uDq/537H6HUfs/q/tH7O7Q6OcZ9Pthk8ohp1EfW24 P2ieXH/4b03of9VfiHRdJ/O9D/h8/T+j+G/+Eeh6HUYw6OXS9JDLDD/4R0+WUP5kf+K9fJ+zY/T/ APSr/iei/Zs+f/uf/Bno6jozh39PhhkMI4cGMzxx2+rDFmn1P4ni6XJl/k5eqz9L1Preh/4R6/8A P6d9P8GxeniltjkhAylLHDPthl83930+i6WGHD0HS+v637Pg/wDG9R/v3g6XpjlPUZNuUT9U9Rhg d/TYs/wfs2P0sv8A4J6ufN0/97Pg/aOn/kdT/IePpenzwxzjHFMQkOnyZMUcfoRyYsRyYvxP8PxY 8uT18vV+l/xH7Rn/AGj8R9LN/wCi3UdP+0dUukEp48nT9PKBlHPHDvjH/wAFzZP2fqeh63qceTJm w9B0uDN+25+nwYP7H/ov0/7V1H7O88ulkcEhHDOMvQlj6qOw/wDhHWfy8XQ5MfpZP/D8v7X+09b/ ANZ4fX/8Hy/+E9T/ALjBpLoBjy4J4MU6iMc/SyQM8XqyyY8vUep1Pqev+FfiGH0vX6jN/wCEYOp9 HD038/8A4fqPrFVUKlUPHj/4zL/4rp//AHZ+KPaqv//U+zfL6nD1cc5zdMcdShDHMZt/+6nny4vT /Zv/AGdyf+a2f/Ur/wCm/wD6HTf4oO2D6c3/AMDRu/FP4en/AOVn/wDrdN/in8PT/wDLzf8A1st/ in8PT/8ALzf/AFsu78U/h6f/AM2Zv/rZO78T/h6f/wA2Zv8A62Xd+J/w4P8AzZm/+tkbvxT+Hp// ADZm/wDrZO78T/h6f/zZm/8ArZ5/2frTlGc48HqgGEZRy5/g/wC79L9n9HL/AOZv/KP/AHD0bvxP +HB/5szf/Wyd34l/Bg/82Zv/AKzXd+JfwYP/ADZm/wDrNd34n/Dg/wDNmb/6zXd+JfwYP/Nmb/6z Xd+JfwYP/Nmb/wCs13fiX8GD/wA2Zv8A6zXd+JfwYP8AzZm/+s13fiX8GD/zZm/+s13fiX8GD/zZ m/8ArNd34l/Bg/8ANmb/AOs07vxL+DB/5szf/Wa7vxL+DB/5szf/AFmu/wDEv4MH/m3N/wDWSif4 j/Bg/wDNub/6yXd+JfwYP/Nmb/6zXf8AiX8GD/zZm/8ArJd/4l/Bg/8ANub/AOsl3/iX8GD/AM25 v/rJd/4l/Bg/825v/rJ5c3T9VnnDLkxYZTx2cR9XP5JT2erk/wCD9PJ/bx/3f7D1bvxL+DB/5tzf /WS7/wAS/gwf+bc3/wBZLv8AxL+DB/5szf8A1ku78S/gwf8AmzN/9Zru/Ev4MH/mzN/9Zru/Ev4M H/mzN/8AWaN34l/Bg/8ANmb/AOs13fiX8GD/AM2Zv/rNd34l/Bg/82Zv/rNd34l/Bg/82Zv/AKzX d+JfwYP/ADZm/wDrNG78T/hwf+bM3/1mu78T/hwf+bM3/wBZru/E/wCHB/5szf8A1mnd+JfwYP8A zZm/+s0bvxP+HB/5szf/AFs30mLqBlnm6nYDKOPHH0ZTn/al1WT+b62PB/6W/wDoJ9FVf//V+zVV VCVVVVVVVVVVVVVVVVVVVVVVCVVCVVVVVVX/2Q== ------=_NextPart_000_0000_01C619F5.950D5A40 Content-Type: multipart/alternative; boundary="----=_NextPart_001_0003_01C619F5.950D5A40" ------=_NextPart_001_0003_01C619F5.950D5A40 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Content-Location: http://www.law.nyu.edu/journals/envtllaw/issues/vol3/2/3nyuelj469.html NYU ELJ ARTICLE (1995) <BODY BGCOLOR=3DWHITE> <BR>=20 <A HREF=3D"3nyuelj469t.html">Please follow this link if your browser = does not support frames.</A> <BR> <BR> <BR> </BODY> ------=_NextPart_001_0003_01C619F5.950D5A40 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Content-Location: http://www.law.nyu.edu/journals/envtllaw/issues/vol3/2/3nyuelj469t.html AN ENVIRONMENTAL JUSTICE CRITIQUE OF RISK = ASSESSMENT NYU LAW = SCHOOL=20 WEB

=A9 New York University = Environmental Law=20 Journal, 1995
Student = Risk=20 Assessment Article
AN ENVIRONMENTAL JUSTICE CRITIQUE OF = RISK=20 ASSESSMENT
Brian D. = Israel =86 =
3nyuelj469=20 

 



 Introduction =
.............................................................. 470

   I.  The Mechanics of Risk Assessment as a Regulatory Tool =
............... 474=20

       A. Regulating Against Risk =
.......................................... 474=20

         1. Definition and History of Risk Assessment =
...................... 475=20

         2. The Function of Risk Assessment =
................................ 478=20

         3. The Risk Assessment Debate =
..................................... 479=20

       B. Risk Assessment Guidelines =
....................................... 482=20

         1. Hazard Identification =
.......................................... 483=20

         2. Dose/Response Analysis =
......................................... 484=20

         3. Exposure Assessment =
............................................ 487=20

         4. Risk Characterization =
.......................................... 488=20

       C. Selecting Hazards =
................................................ 489=20

  II.  Risk Assessment and Theories of Disproportionate Risk =
............... 491=20

       A. Distortions Resulting from Increased Exposure =
.................... 494=20

         1. Multiple Exposures =
............................................. 496=20

         2. Synergistic Effects =
............................................ 497=20

         3. Increased Exposure to Single Substances =
........................ 500=20

         4. Other Exposure Problems =
........................................ 502=20

       B. Distortions Resulting from Increased Susceptibility =
.............. 503=20

         1. Genetic Characteristics =
........................................ 504=20

         2. Disease Frequencies =
............................................ 506

         3. Social Inequalities =
............................................ 507=20

         4. Lifestyle Factors =
.............................................. 508=20

 III.  Possible Regulatory Responses to the Environmental Justice =
Critique=20

         of Risk Assessment =
................................................ 509

       A. Abandoning Risk Assessment =
....................................... 510=20

       B. Correcting the Informational Bias =
................................ 511=20

         1. Information Gathering =
.......................................... 511=20

         2. Generalized Data =
............................................... 513=20

         3. Public Participation =
........................................... 516=20

         4. Burden of Proof =
................................................ 519=20

         5. Cleanup Standards =
.............................................. 520=20

 Conclusion =
................................................................ 521
*470

Introduction

   =  =20 The environmental justice movement seeks to ensure that = environmental=20 protection is as much a commitment in poor communities as it is in = wealthy=20 communities and as much a commitment in communities of color as it = is in=20 white communities. 1

    =20 Environmental justice begins with the troubling observation that=20 environmentally hazardous facilities are disproportionately = distributed in=20 our society on the basis of race and class. 2

*471

 &n= bsp;  =20 However, environmental justice is not limited to this claim. = Environmental=20 justice is equally concerned with the government's failure to = adequately=20 protect public health and the environment. As stated by Dr. Robert = Bullard, one of the leading scientists in the environmental = justice=20 community, "[n]ot only are people of color differentially impacted = by . .=20 . pollution, they can expect different treatment by the = government." 3

    =20 Concern about government action often focuses on the allegedly = uneven=20 enforcement of environmental laws. 4 In the Superfund context, = for=20 instance, one national study indicates that cleanups of hazardous = waste=20 sites are slower, 5 and remedies chosen = *472 less=20 permanent, 6 in minority communities = than in white=20 communities. In other contexts, fines for violations of federal = hazardous=20 waste laws are up to 500 times greater in white communities than = in=20 minority communities. 7

    =20 Independent of inequitable enforcement issues, environmental = justice=20 activists also argue that government health studies fail to detect = adverse=20 health effects that poor and minority communities experience. 8 In particular, = environmental justice=20 activists criticize the regulatory methodology used to assess the = type and=20 degree of health risk posed to the public by a particular = substance. This=20 methodology, termed risk assessment, is an increasingly = fundamental aspect=20 of environmental and public health regulation. 9 To the degree that risk = assessment is=20 a requisite element for regulatory action, therefore, a risk = assessment=20 methodology *473 that obscures risks on the basis of race = and class=20 results in less than adequate environmental and health protection = for=20 members of these groups. 10

     = The=20 idea of government health studies failing to notice health effects = in=20 particular communities is alarming for two reasons. First, it = indicates=20 that the government may be relying on research methodologies that = preclude=20 the possibility of equal protection from pollution. This is true = because=20 distortions in risk assessment may result in a situation in which = the same=20 government regulation leads to adequate protection for one group = and=20 inadequate protection for another group. 11 Second, objective = scientific=20 research by the government is typically the aspect of regulation = thought=20 to be least susceptible to demographic or political criteria. 12

     = Part=20 I of this Student Article defines risk assessment and discusses = the=20 mechanics of risk assessment as a regulatory tool. In so doing, it = highlights the aspects of risk assessment that are particularly = relevant=20 to an environmental justice=20 critique.

*474

     Part = II=20 explores the claim that risk assessment fails to detect risks in=20 particular communities by detailing the informational biases = contained in=20 the practice of risk assessment. Such informational biases, if = they exist,=20 systematically distort or hide the health risks in communities of = color=20 and poor communities. These alleged biases tend to fall into one = of two=20 broad categories. First, risk assessment fails to adequately = account for=20 above average exposures. Second, risk assessment fails to = adequately=20 account for above average susceptibilities. Part II of this = Student=20 Article discusses a number of risk assessment failures within each = of=20 these two categories and reviews some of the scientific literature = discussing these problems.

     Part = III=20 suggests changes and innovative approaches to facilitate the = improved use=20 of risk assessment methodologies for highly exposed or highly = susceptible=20 populations. In addition, Part III examines some of the regulatory = obstacles to making risk assessment more = equitable.


I. The=20 Mechanics of Risk Assessment as a Regulatory Tool

A. = Regulating Against Risk

     = Congress has=20 charged many federal agencies with the responsibility for = determining the=20 risk posed by a multitude of substances present in our food, 13 air, 14 water, 15 soil, 16 work places, 17 *475 homes, 18 and products. 19 Frequently, Congress = grants these=20 agencies broad authority to regulate dangerous substances or = pollutants in=20 an effort to reduce or eliminate the risk of harm. 20

1. Definition = and History=20 of Risk Assessment

     The = determination=20 of the specific risk posed by exposure to a particular substance = or hazard=20 is called "risk assessment." Risk assessment is a = multidisciplinary=20 methodology involving, for example, toxicology, genetics, = immunology,=20 pharmacokinetics, 21 and epidemiology. Its goal = is to=20 identify and measure the risk that a particular substance presents = at a=20 given exposure level. Specifically, the Department of Health and = Human=20 Services defines *476 risk assessment as the "qualitative = or=20 quantitative estimation of the likelihood of adverse effects that = may=20 result from exposure to specified health hazards or as a result of = the=20 absence of beneficial influences." 22

    =20 Quantitatively, risk assessment methodology yields a number used = to assist=20 the regulatory body in the promulgation of rules and regulations. = The=20 Environmental Protection Agency's (EPA) definition of risk = assessment=20 highlights this quantitative focus: "[r]isk assessment = characterizes the=20 likelihood of a chemical agent or mixture to cause an adverse = health=20 effect for humans and on a case-by-case basis provides a numerical = way to=20 gauge the possible impact on a population(s) if exposure were to = occur."=20 23 The final number = represents the=20 probability that a particular risk (e.g., cancer) will occur at = certain=20 exposure levels. For example, according to the New York State = Department=20 of Health, there would be between 100 and 150 excess cancer deaths = per one=20 million people as a result of grain consumption if ethylene = dibromide were=20 regulated at a maximum exposure level of 30 parts per billion. 24

    =20 While risk assessment has been a tool of agency decision making = since the=20 1970s, 25 the technique became = dramatically=20 more pervasive and sophisticated following the Supreme Court's = benzene=20 decision in 1980. 26 Benzene, a substance used = in the=20 petrochemical industry, is associated with adverse health effects = such as=20 leukemia, chromosomal alterations, and nonmalignant blood = disorders. 27

*477

 &= nbsp;  =20 In 1977, after significant public pressure, 28 the Occupational Safety = and Health=20 Administration (OSHA) published a proposed permanent standard for=20 controlling occupational exposure to benzene. 29 Industry groups = immediately=20 challenged OSHA's standard, and, in 1979, the litigation reached = the=20 United States Supreme Court. 30 The issue for the Supreme = Court to=20 decide was whether the Court of Appeals correctly set aside OSHA's = standard on the ground that it was unsupported by appropriate = findings. 31 OSHA argued that utilizing = risk=20 assessment was inappropriate due to a lack of sufficient data and = the fact=20 that such an exercise would require the government to place a = dollar=20 amount on human life. 32 The Supreme Court held = that because=20 the relevant statute required standards "reasonably necessary . . = . to=20 provide safe or healthful employment," 33 the statute obligated the = government=20 to make threshold findings that the workplace is unsafe. 34

    =20 Furthermore, the Supreme Court held that OSHA did not meet its = burden of=20 proof that benzene presented a significant risk in the workplace. = 35 The Court concluded that = "[a]lthough=20 the Agency ha[d] no duty to calculate the exact probability of = harm, it=20 d[id] have an obligation to find that a significant risk [wa]s = present."=20 36 Although the Supreme Court = was=20 careful not to demand scientific certainty or excessive = mathematical=20 burdens, 37 the effect of the decision = was to=20 establish quantitative risk assessment as a regulatory = prerequisite. 38

*478

2. = The=20 Function of Risk Assessment

    =20 Quantitative risk assessment performs one of two regulatory = functions. 39 First, it establishes safe = levels of=20 exposure in order to determine proper regulatory standards for = particular=20 substances. 40 Second, it has a = comparative=20 function: an agency may quantify and compare risks in order to = make=20 allocative decisions about which substance or site to regulate = first.=20 EPA's Superfund program provides a prototypical example of = comparative=20 risk assessment because it incorporates a highly formalized risk=20 comparison mechanism to create its National Priority List (NPL). = 41 This risk comparison, = called the=20 Hazards Ranking System (HRS), 42 incorporates risk = assessment=20 techniques to "accurately [assess] the relative degree of risk to = human=20 health and the environment posed by sites and facilities." 43

    =20 Superimposed upon the standard-setting and comparative allocations = function is the distinction between prospective and retrospective = risk=20 assessments. 44 Prospective risk = assessments guide=20 some future action. For example, EPA may determine an appropriate = amount=20 of future emissions of a substance. Similarly, through the use of=20 comparative risk assessments, EPA may determine that a certain = waste=20 facility is a priority and must be remedied. Conversely, = retrospective=20 risk assessments are incorporated by an agency after some initial=20 regulatory commitment or determination of the need for action. For = example, retrospective *479 risk assessments are used as = part of=20 the remedy selection and cleanup standard determination at a = Superfund=20 site. 45

3. The Risk = Assessment=20 Debate

     Risk assessment is = pervasively=20 controversial. Legislative debates concerning environmental = protection,=20 particularly current debates, often stall on the question of risk=20 assessment's proper role within our regulatory agencies. 46 There are at least three = major=20 controversies that frequently resurface. First, many observers = fear that=20 risk assessment is an obstacle to regulatory action. 47 These observers argue that = because=20 risk assessment can never account for inevitable scientific = uncertainty, a=20 regulatory scheme that requires the demonstration of a risk before = authorizing a response will be immobile in the face of perceived, = though=20 unproven, risks. 48 Similarly, some observers = argue that=20 mandated risk assessment undermines agency flexibility by = establishing=20 absolute quantitative criteria. 49 Other observers answer = *480=20 these criticisms by arguing that risk assessment is critical in = order to=20 ensure that the government regulates the correct substances to the = proper=20 degree. 50

    =20 Second, because risk assessment is ultimately an allocative = device,=20 ethical questions arise. 51 As discussed above, = comparative risk=20 assessment is the quantitative process of comparing the results = from=20 individual risk assessments to decide which hazards deserve = regulatory=20 priority. 52 For those who argue that = the=20 government has an absolute responsibility to eliminate industrial = and=20 commercial risks, it is irresponsible to use risk assessment as a=20 comparative allocative tool because it postpones the elimination = of lower-=20 priority risks. 53 On the other hand, for = those who=20 argue that it is the ethical responsibility of government not to = provide=20 absolute protection but rather to act efficiently and = cost-effectively,=20 the use of risk assessment is required. 54 This is true because = *481 it=20 allows government to regulate efficiently by permitting regulators = to=20 determine which hazards constitute the greatest threat and which = hazards=20 can be reduced cost-effectively.

     = The third=20 major controversy surrounding risk assessment involves its = scientific=20 methodologies. Nearly every scientific aspect of risk assessment = -- the=20 determination of causation, the extrapolation of health responses = caused=20 from high doses of a substance to health responses caused from low = doses,=20 the extrapolation from animal responses to human responses, and = the=20 estimation of exposure levels -- is controversial. 55 It is interesting to note = that those=20 observers advocating for more stringent standards and those = advocating for=20 less stringent standards both share a frustration with the = "scientific=20 failings" of risk assessment. Those arguing for more stringent = standards=20 contend that the science of risk assessment excludes critical = information,=20 such as background exposure levels and access to health care. 56 Those arguing for less = stringent=20 standards (alternatively stated as more efficient regulation) = attack the=20 overly conservative assumptions that risk assessment employs and = that=20 allegedly exaggerate the real risk of exposure. 57

*482

B. = Risk=20 Assessment Guidelines

     In 1983, = in an=20 effort to standardize and improve the government's use of risk = assessment,=20 the National Research Council of the National Academy of Sciences = (NAS)=20 published Risk Assessment in the Federal Government: Managing the = Process.=20 58 NAS endeavored to make the = risk=20 assessment process more objective by creating a bright-line = distinction=20 between risk assessment and risk management. 59 Risk management would = consist of=20 difficult policy questions, while risk assessment would consist = only of=20 objective scientific techniques. 60 NAS further proposed that = federal=20 regulatory agencies adopt risk assessment guidelines that would=20 incorporate four basic stages: hazard identification; = dose-response=20 assessment; exposure assessment; and risk characterization. 61 In response to the NAS = book, EPA=20 implemented and published a set of its own guidelines to assist = its=20 scientists in using NAS's four-pronged approach to risk = assessment. 62

*483

1.=20 Hazard Identification

     Most = risk=20 assessments begin with an examination of the health effect of a = single=20 substance, as opposed to a combination of substances or chemical = mixtures.=20 63 Although single substance = risk=20 assessments are inherently problematic, 64 a number of arguments = support this=20 approach.

     First, single substance = studies=20 permit easier generalization because they allow regulators to = refer to a=20 commonly encountered substance as opposed to a rarely encountered = mixture.=20 Second, single substance risk assessments are less expensive = because they=20 involve fewer basic substances than complex mixtures. Third, = single=20 substance risk assessments are, in some cases, required by = legislation. 65

     = In=20 the context of environmental regulation, hazard identification is=20 characterized by a qualitative judgment that exposure to a = substance=20 causes a disease or other health effect. While a particular = substance may=20 be responsible for a number of health effects, hazard = identification only=20 focuses on one or very few such effects. 66 These effects are called = "endpoints"=20 and may encompass a wide variety of effects including cancer, = reproductive=20 abnormalities, developmental disorders, central nervous system = symptoms,=20 trauma, infections, and rashes. 67 Cancer is often chosen as = the=20 endpoint for a given risk assessment because it is easy = *484 to=20 identify, 68 it demands the bulk of = public=20 concern, 69 and it tends to be the = most=20 sensitive endpoint making it a safety net for other possible = endpoints. 70 To the extent that = carcinogenesis is=20 not the most sensitive endpoint of a particular substance, = however, risk=20 assessments that rely on the identification of cancer may fail to = notice=20 other adverse health effects. 71

    =20 Though hazard identification consists only of the qualitative=20 determination of causation, it is nonetheless an exhaustive = procedure. 72 Upon reviewing all of the = evidence,=20 an agency then evaluates and interprets the weight of the evidence = --=20 often with guidelines 73 -- to conclude whether or = not an=20 adverse health effect exists as a result of the = substance.

2.=20 Dose/Response Analysis

     The = second=20 stage of risk assessment is called hazard characterization or=20 dose/response analysis. This stage attempts to quantify = *485 and=20 clarify the qualitative results from the hazard identification by=20 anticipating the expected response caused by a range of exposures. = In the=20 case of "threshold" hazards, where there is a threshold amount = below which=20 there is no adverse health response, risk assessors attempt to = identify=20 the no observed effects level (NOEL) or the lowest observed = effects level=20 (LOEL). 74 Subsequent dose/response = curves then=20 measure the severity of health responses at various exposure = levels. 75

     = In=20 the case of "nonthreshold" hazards, where there is no known = exposure level=20 below which there is no adverse effect, the hazard = characterization stage=20 of risk assessment attempts to quantify the probability of = experiencing=20 the adverse effect at various exposure levels. 76

     = For=20 purposes of an environmental justice critique of risk assessment, = two=20 aspects of the dose/response analysis are important to note. = First,=20 dose/response analysis requires that the risk assessor extrapolate = from=20 initial data at a particular dose level to other dose levels. 77 For reasons of cost and=20 observability, initial toxicological studies are often performed = at high=20 doses. 78 Thus, risk assessors = attempt to=20 extrapolate from high dose responses to low dose responses. They = have=20 developed numerous methods of extrapolation of varying degrees of=20 sophistication. 79 The most prevalent tool in = extrapolation is an assumption of linearity, such that there is an = imagined straight line that correlates high dose/high responses to = low=20 dose/low responses. 80 Such a method carries with = it=20 obvious uncertainties for individuals exposed to low doses. 81

*486

 &= nbsp;  =20 Second, independent of the dose/response extrapolations, hazard=20 characterization must account for differing sensitivities and=20 susceptibilities among the exposed population. 82 Although the EPA = Guidelines for=20 Carcinogen Risk Assessment (Carcinogen Guidelines) acknowledge the = fact=20 that "human populations are variable with respect to genetic = constitution,=20 diet, occupational and home environment, activity patterns, and = other=20 cultural factors," 83 the Carcinogen Guidelines = present no=20 other advice or information to assist risk assessors in accounting = for=20 variables that influence susceptibility. Currently, no generally = accepted=20 or systematic method exists to tailor risk assessments to address = subgroup=20 sensitivities. 84 As a result, most risk = assessments=20 account for very little, if any, special susceptibilities. 85 EPA has acknowledged this = in its=20 Environmental Equity report:

     = Information=20 from epidemiologic studies has largely involved studies on the = standing=20 work force, i.e. exposures to healthy White male workers. Given = this=20 limited study population, the extrapolated hazards to the general=20 population may not portray the range of consequences to children, = the=20 elderly, the sick and infirmed, females, racial/ethnic groups, or=20 low-income populations. 86

*487

3.=20 Exposure Assessment

     Exposure=20 assessment consists of estimating the number of people exposed to = a=20 certain chemical substance and the amount of exposure to that = substance.=20 87 In 1992, EPA published = Guidelines=20 for Exposure Assessment, which "discuss and reference a number of=20 approaches and tools for exposure assessment." 88

     = As=20 will be discussed more fully below, 89 there are at least two = points of=20 contact between the exposure assessment methodology and an = environmental=20 justice critique of risk assessment. First, because exposure = assessment=20 attempts to predict rather than actually measure exposure to a = chemical=20 substance, it necessarily involves generalization. In its = Environmental=20 Equity report, EPA acknowledged the problem of generalizing = exposure:
[D]emographic categories may be useful markers for=20 identifying population subgroups that have some likelihood of=20 experiencing exposures significantly different from the average = exposure=20 and, thereby, possibly different health risks from the average=20 population. 90
An = environmental justice=20 critique of risk assessment, then, inquires as to whether or not=20 particular subpopulations consistently experience exposure at = higher than=20 average levels. 91=20 *488

     Second, exposure = assessments=20 tend not to examine multiple and indirect pathways and multiple = and=20 indirect sources of possible exposures. 92 Overlooked exposure routes = or=20 sources include exposure to air pollutants through food, = automobile=20 exhaust, and indoor air. 93 To the degree that = particular=20 subgroups are more heavily exposed to these indirect exposures, an = environmental justice critique posits that risk assessment fails = to=20 adequately protect those groups. For example, one study indicates = that=20 low- income, mostly African American residents of Harlem and = Washington=20 Heights in New York City have higher than average indoor exposure = to=20 nitrogen dioxide and carbon monoxide, possibly caused by the use = of=20 unvented gas-fired kitchen stoves as a source of heat. 94

4. Risk=20 Characterization

     The final = stage of=20 risk assessment is called risk characterization. Very simply, risk = characterization is the incorporation of data from the dose = response and=20 exposure assessment stages into a quantified, communicated risk=20 probability. An environmental justice critique of risk assessment = finds=20 fault with the risk characterization stage to the degree that this = stage=20 incorporates data from the already flawed prior=20 parts.

     According to EPA = guidelines, risk=20 assessors must identify and explain all uncertainties and = assumptions=20 inherent in the results. 95 Nonetheless, risk = assessments vary=20 in terms of explicit reference to subgroup susceptibilities and = subgroup=20 exposures. 96 Especially for assessments = used to=20 set national standards, risk characterizations refer to the = average=20 individual exposed to the average dose. 97

*489

C.=20 Selecting Hazards

     An important = problem=20 surfaces regarding the initial decision to perform a risk = assessment=20 analysis: how and by whom is the decision made to initiate the = study?=20 There are over 65,000 chemicals currently in use, 98 and people may be exposed = to=20 millions of unique combinations of those 65,000. Clearly, the = government=20 is unable to perform assessments of the risk posed by all such = substances=20 and combinations, and so it becomes necessary to ask how = government=20 agencies choose which assessments to perform. The absence of = information=20 regarding the assessment initiation process stands in stark = contrast to=20 the extensive literature surrounding the actual assessment = methodology. A=20 constructive environmental justice response to risk assessment = problems=20 must address this crucial stage in risk=20 regulation.

     There are a number of = paths,=20 both within the government and external to the government, that = may lead=20 to the initiation of a risk assessment. Internal methods include=20 requirements provided by Congress. For example, the Toxic = Substances=20 Control Act (TSCA) establishes an interagency committee charged = with the=20 task of recommending to EPA chemicals or chemical groups for = priority=20 testing consideration. 99 In making its = recommendations, this=20 committee must consider a number of factors and give priority to = "those=20 chemical substances and mixtures which are known to cause or = contribute to=20 or which are suspected of causing or contributing to cancer, gene=20 mutations, or birth defects." 100

     = External contributions to the decision to initiate a risk = assessment=20 involve observations by persons outside of the government = communicated to=20 a responsible agency. In 1979, for example, consumer and worker = complaints=20 regarding formaldehyde led directly to the establishment of the=20 Interagency Regulatory Liaison Group to study the compound. This = pressure=20 ultimately led to both a number of risk assessments performed = by=20 *490 different agencies and regulatory decisions promulgated = by EPA,=20 the Consumer Product Safety Commission, and OSHA. 101 Similarly, risk = assessments and the=20 ultimate ban on the flame retardant, Tris phosphate, resulted = entirely=20 from petitions by independent scientists and environmental groups = such as=20 the Environmental Defense Fund (EDF). 102 These external petitions = were=20 directed to the Consumer Product Safety Commission authorized to = regulate=20 under the Hazardous Substance Control Act. The case of ethylene = dibromide,=20 a highly toxic substance used primarily as a grain and produce = fumigant,=20 also provides an example of external contributions to government = risk=20 assessments. Pressure by EDF and labor unions led to accelerated = risk=20 assessments and ultimate regulations on the part of EPA, OSHA, and = the=20 National Institute for Occupational Safety and Health (NIOSH). 103

     = The=20 existence of internal and external paths to the initiation of a = risk=20 assessment suggests a number of conclusions. First, initial = suspicion of a=20 substance or combination is a critical part of risk assessment and = ultimately regulation. There is no systematic attempt to assess = all=20 possible chemical risks, and our knowledge is limited to the = extent that=20 we suspect, and then analyze, particular substances. Furthermore, = absent=20 congressional requirements to study particular substances, it is = often=20 unclear how an agency becomes initially aware of a substance or = mixture so=20 as to determine whether an assessment is necessary. Finally, = organizations=20 and individuals external to the government frequently contribute = to the=20 decision to initiate a risk assessment. These observations lay the = foundation for the recommendation that risk assessment become more = attentive to public concerns. As Part III discusses, 104 increased public = participation in=20 the risk assessment process is necessary to alleviate some of the = possible=20 informational biases discussed in the next=20 part.

*491

II. Risk Assessment and = Theories of=20 Disproportionate Risk

     = Environmental=20 justice activists believe that government health studies = systematically=20 undervalue the environmental health threats experienced by poor=20 communities and communities of color. Environmental justice = activists and=20 scholars voice four objections to the risk assessment process: (1) = health=20 studies and risk assessments are used to justify government = inaction in=20 the face of real risks; 105 (2) risk management = decisionmak-=20 ing is politically motivated, ignoring risks experienced in poor = or=20 minority communities; 106 (3) scientific risk = assess- ments=20 fail to account for community perceptions of risk; 107*492 and (4) the = risk=20 assessment methodology is itself discrimina- tory. 108

     = This Part examines the theoretical support for the last objection. = This=20 objection suggests that there exists an informational bias within = risk=20 assessment as performed by regulatory agencies. The informational = bias=20 results from the failure of risk assessors to incorporate a number = of=20 critical variables, including information regarding unusual = exposure=20 patterns and unusual susceptibilities. As a result, environmental = justice=20 activists assert that risk assessments tend to distort or hide = health=20 effects in communities of color and poor communities, thereby = leading to=20 less than adequate environmental protection for those groups. 109

*493

 =    =20 There are a number of theoretical reasons why risk assessment may=20 methodologically fail to detect health effects in poor communities = and=20 communities of color. 110 It is useful to separate = these=20 reasons into two broad categories: (A) failures of risk assessment = that=20 disproportionately affect poor and minority communities because = these=20 communities are more likely to be exposed to risk; and (B) = failures of=20 risk assessment that disproportionately affect poor and minority=20 communities because these communities are more likely to be = susceptible to=20 risk. 111

     = The=20 following table outlines these two categories of reasons and lists = examples of the phenomena that risk assessment currently fails to = observe.=20 The fact that risk assessment fails to observe these phenomena = creates the=20 theoretical underpinnings of an environmental justice critique. 112



*494

A. = Distortions=20 Resulting from Increased = Exposure

     Risk=20 assessment involves a number of exposure assumptions that dilute = the=20 significance of assessment results. For instance, calculations = often do=20 not account for the dangers associated with multiple exposures, 113 chemical mixtures, 114 above-average exposure, = 115 and long-term, small-dose = exposures. 116

     = While each of these problems affects the population in general, to = the=20 degree that certain subgroups are more highly exposed to regulated = substances, those groups suffer disproportionately from failures = of risk=20 assessment to adequately incorporate exposure realities. There is = already=20 substantial evidence that poor and minority communities are=20 disproportionately exposed to regulated substances such as air = pollutants,=20 117 *495 water = pollutants, 118 hazardous wastes, 119 and pesticides. 120 Furthermore, most studies = show that=20 race, independent of economic status, is the single most reliable=20 predictor of exposure to pollution. 121

     = There are at least three exposure-related examples of how risk = assessment=20 may fail to detect health effects: (1) multiple exposures; (2) = synergistic=20 effects; and (3) increased exposure to single substances. = Furthermore,=20 other uncertainties within risk assessment, including those = created by the=20 bias toward observing *496 cancer endpoints and the = tendency to=20 examine only short-term effects of high doses, are aggravated by = increased=20 exposure.

1. Multiple=20 Exposures

     Risk assessment = determines=20 the probability that a person exposed to a particular substance = will=20 experience a given health effect. The single substance = methodology,=20 however, fails to account for multiple exposures. 122 Therefore, the acceptable = risk=20 level is never as conservative as it appears because people are = never=20 exposed to only one substance. 123 Imagine an EPA acceptable = risk=20 level of one in one million (10 -6']) chances of death. Faced with = this=20 scenario, EPA might set emission standards, cleanup standards, and = technology standards such that each substance would expose people = to no=20 more than the amount which would cause death in one person out of = one=20 million.

     Although this is a = seemingly=20 conservative standard, it is undermined by the potential that = people are=20 exposed to hundreds, if not thousands, of regulated substances. If = person=20 n is exposed to 100 substances at a level that would cause death = in one=20 person per million (10 -6']), n's actual additive risk of death is = one in=20 ten thousand (10 -4']). 124

     = While this distortion is important to the public in general, it is = critical to observe that such a bias may have a disproportionate = effect in=20 poor communities and communities of color where exposure to = multiple=20 substances tends to be higher. For example, Atgeld Gardens is a = huge=20 housing project in Chicago which is home to 150,000 residents, of = whom 70%=20 are African American and 11% are Latino. 125 Surrounding this = community is a=20 "toxic *497 doughnut" 126 in which there are 50 = active or=20 closed commercial hazardous waste landfills, 100 factories, and = 103=20 abandoned toxic waste dumps. 127 Observers note, however, = that "risk=20 assessment data collected by the state of Illinois and the federal = EPA for=20 facility permitting have failed to take into account the = cumulative and=20 synergistic effects of so many sources of poisons in one = community." 128

2. Synergistic=20 Effects

     Because risk = assessments tend=20 to focus on the health effects of single substances, they also = fail to=20 detect those risks that may occur when numerous chemicals = interact. 129 Synergistic or = interactive effects=20 occur when the presence of one component alters the effect of = another. 130 The new substance that = results from=20 the combination of two or more components is called a chemical = mixture. 131 The resulting chemical = mixture may=20 cause an additive response (whereby the two chemicals act = individually),=20 132 an antagonistic response = (whereby=20 one chemical prevents a response of another chemical such that the = total=20 response is less *498 than expected), 133 or a synergistic response = (whereby=20 the two chemicals interact to cause a response that is greater = than the=20 expected response). 134 A synergistic response = may result=20 in one chemical aggravating the toxicity of another chemical 135 or in the creation of a = new=20 compound that is toxic in a manner dissimilar to that of either of = the=20 component chemicals. 136

     = Mixtures result from a number of scenarios. Often, complex = mixtures are=20 the by-products of industrial processes such as coke oven = emissions or=20 diesel exhaust. 137 Sometimes, mixtures are=20 intentionally created commercial products, such as PCBs, gasoline, = and=20 pesticides. 138 Hundreds of chemicals may = also=20 combine when disposed of at a common waste site. 139 Although not mentioned in = EPA's=20 chemical mixture guidelines, a fourth scenario for the creation of = chemical mixtures is the interaction of air emissions. Fifth, the=20 interaction of substances that have been released into a body of = water can=20 create chemical mixtures. 140 Finally, mixtures also = include the=20 possible synergistic effects that may occur when an individual is=20 separately exposed to individual substances such that the mixture = occurs=20 within the individual.

     While there = is a=20 "paucity of empirical data" regarding the health effects of most = chemical=20 mixtures, 141 the regulatory importance = of=20 considering synergistic effects is widely acknowledged. 142 *499 For example, = the=20 statutory mandate of the Agency for Toxic Substances and Disease = Registry=20 (ATSDR) requires that ATSDR "[w]here feasible . . . develop = methods to=20 determine the health effects of such substance in combination with = other=20 substances with which it is commonly found." 143 Likewise, EPA, in its = Guidelines=20 for the Health Risk Assessment of Chemical Mixtures (Mixture = Guidelines),=20 observed that "[w]hile some potential environmental hazards = involve=20 significant exposure to only a single compound, most instances of=20 environmental contamination involve concurrent or sequential = exposures to=20 a mixture of compounds that may induce similar or dissimilar = effects." 144

     = The=20 Mixture Guidelines are worth examining in more detail both because = they=20 have attracted a fair amount of attention and because they = illustrate some=20 of the difficulties associated with risk assessment. = Unfortunately, the=20 points where the Mixture Guidelines are weakest are the exact = points in=20 which the exposure data linking minority and poor communities are = the=20 strongest.

     The Mixture Guidelines = propose=20 various methods for assessing the risks posed by mixtures. 145 In cases where sufficient = data=20 regarding the specific mixture or a similar mixture are not = available, the=20 Mixture Guidelines require risk assessors to assume additivity. 146 Additivity models apply = simple=20 cumulative assumptions about the effect of multiple substances = such that=20 the effect of a mixture including chemicals "A" and "B" is = considered to=20 be equal to the sum of responses associated with "A" plus the = responses=20 associated with "B." The Mixture Guidelines acknowledge the = limitations of=20 additivity, noting that in cases of synergism, additivity will=20 underestimate risk. 147 Still, because of = *500 the=20 lack of reliable data regarding chemical interactions, additivity=20 assumptions will be incorporated in almost all cases. 148

     = This use of additivity models to estimate risk may have a = disproportionate=20 effect in poor communities and communities of color. As discussed = above,=20 and as acknowledged by EPA, additivity fails to account for the = potential=20 synergistic effects associated with complex mixtures. Furthermore, = the=20 greater the number of component substances, the less reliable are = the=20 additive results. 149 Substantial data = demonstrate that=20 low-income and minority people are significantly more likely than = the rest=20 of society to live near complex mixture "scenarios." 150 These scenarios include = toxic waste=20 and hazardous waste facilities, 151 as well as situations = such as=20 Atgeld Gardens, a community surrounded by multiple sources of = chemical=20 emissions. 152

3. Increased = Exposure to=20 Single Substances

     Quantitative = estimates of risk that do not acknowledge a range of potential = exposures=20 fail to notice demographic correlations. Not only will generalized = exposure assumptions distort risks that are disproportionately=20 distributed, but such assumptions are also inefficient, leading to = regulations that are simultaneously over- and=20 under-inclusive.

     This phenomenon = is=20 exemplified by generalized dietary exposure assumptions. For = example, if=20 pesticide x is used for avocados, an assessment of risk that = assumes that=20 a certain amount of pesticide x is acceptable based on an estimate = that=20 all Americans eat n avocado(s) a year fails to protect those who = eat many=20 more avocados. For the purpose of regulating the intake of = pesticide x,=20 inefficiency results for items that have inverse-bell consumption=20 patterns, such that people eat either many or no avocados. = Clearly, the=20 use of an average becomes increasingly meaningless as more and = more people=20 adopt extreme = behavior.

*501

    =20 Among environmental justice advocates, the most frequently cited = example=20 of risk assessment failure concerns consumption patterns of fish. = A number=20 of studies corroborate the claim that minority populations are=20 significantly more likely to consume fish as a source of = subsistence=20 protein, while non-minority persons are more likely to consume = fish on a=20 more sporadic basis. 153 Furthermore, most risk = assessments=20 assume that the population consumes skinless, trimmed fillets, yet = "[t]he=20 evidence suggests that ethnic minorities are more likely to eat = fish with=20 the skin." 154 This difference is = critical because=20 toxins in fish are concentrated in the skin and fatty tissues, = neither of=20 which are eaten as part of a fillet. 155

     = Scientists at the University of Michigan performed an illustrative = study=20 based on a Michigan rule that regulates the discharge of toxic = chemicals=20 into Michigan's surface waters. 156 The rule assumes an = average fish=20 consumption of 6.5 grams/person/day and regulates accordingly. 157 The study concluded that = the=20 average consumption of fish by Whites was 17.9 grams/day/person, = while the=20 average consumption for African Americans and Native Americans=20 respectively was 20.3 grams/day/person and 24.3 grams/day/person. = 158 Therefore, the study = concluded that=20 not only was the Michigan rule assuming far too low of an exposure = level=20 but also that there are significant exposure differences among = sub-groups=20 within the Michigan population. 159

*502

4. = Other Exposure Problems

     Any=20 characteristic of risk assessment that tends to underestimate the = risk of=20 an adverse health effect will disproportionately affect those who = are=20 disproportionately exposed. The preference for the use of = carcinogenesis=20 as the endpoint for risk assessments provides one example. 160 Cancer is often thought = to be the=20 most sensitive health effect and therefore an appropriate endpoint = to=20 screen for all possible health effects. However, other health = effects,=20 such as developmental or reproductive problems, frequently result = without=20 carcinogenesis and may tend to go unanalyzed. 161 While all persons exposed = are at=20 risk, this practice most profoundly affects those who are=20 disproportionately exposed. 162

     = Similarly, risk assessments tend to focus on short term exposures = to high=20 doses. The dose/response stage of risk assessment then = extrapolates to=20 lower doses over longer durations. Again, to the degree that these = extrapolations fail to accurately detect the effect of low-dose = exposures=20 over a long term (e.g., those experienced by people living near an = industrial facility), those who are most exposed to that pattern = will be=20 underprotected. 163

     = This Part has demonstrated some of the possible informational = distortions=20 or biases within risk assessment as they relate to exposure. = Single=20 substance risk assessments do not account for the cumulative = effects of=20 multiple exposures, nor do they account for the synergistic = effects of=20 exposures to chemical mixtures. Risk assessments generally = incorporate=20 exposure assumptions, creating the possible distortion of exposure = patterns that exceed the assumed exposure value. To the degree = that poor=20 people and communities of color are disproportionately exposed to = chemical=20 substances, these biases will result in environmental *503=20 protection for those groups which is less than is necessary to = adequately=20 safeguard health.

B. Distortions Resulting from = Increased=20 Susceptibility

     While an = analysis of=20 subgroup susceptibility is inherent in much of the statutory and = judicial=20 law, 164 risk assessments have = generally=20 failed to examine susceptibility as a function of race, ethnicity, = and=20 income. 165 For example, = epidemiologic studies=20 used in regulatory risk assessments involve studies on the general = work=20 force, involving mostly healthy white males. 166 To the degree that = subgroup=20 susceptibilities are considered, health assessments traditionally = examine=20 age, gender, and a handful of disease groups such as asthmatics. = 167 Unfortunately, such = assessments=20 tend not to provide any racial or economic examples of subgroup=20 considerations. 168

     = Yet, there is substantial evidence suggesting relevant = susceptibilities in=20 minority and poor communities. 169 As discussed below, = *504=20 group susceptibility to pollutants may be affected by the = following: (1)=20 genetic characteristics; (2) disease frequencies; (3) social = inequalities=20 including access to health care, unemployment rates, and = educational=20 levels; and (4) lifestyle factors such as substance abuse and = nutritional=20 deficiencies. 170

1. Genetic=20 Characteristics

     A high = percentage of=20 certain subgroups have a genetic red blood cell deficiency related = to the=20 enzyme, glucose- 6 -phosphate dehydrogenase (G- 6 - PD). 171 It is found in 16% of = African=20 American men, 12-13% of Filipinos, and 11% of Mediterranean Jews. = 172 The G- 6 -PD deficiency = is=20 associated with increased susceptibility to the toxic effects of = ozone,=20 copper, chlorite, and other environmental and occupational = pollutants. 173 Edward Calabrese refers = to this red=20 blood cell deficiency as a paradigmatic example of the = relationship=20 between risk regulation and susceptible subgroups. 174 He argues that this = combination of=20 subgroup susceptibility with an environmental pollutant creates a=20 situation in which "a high level of environmental stressor (ozone) = is in=20 an area inhabited by those least capable of sustaining suitable=20 homeostatic-compensatory responses to offset the stress." 175 Calabrese also notes the = absence of=20 adequate quantitative risk assessments and the "nearly total lack = of=20 epidemiologic studies designed to assess the impact of industrial = and=20 environmental oxidants in G- 6 -PD deficients." 176

     = Other genetic characteristics correlated to ethnicity may increase = susceptibility to environmental exposures. Sickle-cell anemia = *505=20 and sickle-cell trait, which affect almost exclusively African = Americans,=20 may increase susceptibility to benzene, lead, cadmium, and carbon=20 monoxide, among other compounds. 177 Alcohol dehydrogenase = variance,=20 estimated to occur in 70% of the Japanese population and 20% of = the Swiss=20 population, is associated with a faster than normal metabolization = of=20 alcohol-related substances, such as ethanol. 178 Oculocutaneous albinism, = observed=20 at very high frequency among Native Americans (e.g., 1 in 227 = among the=20 Hopi), is thought to increase susceptibility to the toxic effects = of solar=20 radiation. 179

     = Furthermore, repeated exposure to certain toxicants may cause = genetic=20 characteristics that increase susceptibility. While little is = known about=20 this condition, sometimes called "chemically induced = hypersensitivity," 180 some observers suggest = that=20 minority populations are more likely to be exposed to the = particular=20 toxicants associated with causing the disorder. 181

*506

2. = Disease Frequencies

     Disease=20 frequencies differ according to racial, ethnic, and socioeconomic = factors=20 182 and may affect = susceptibility to=20 the effects of exposure to environmental contaminants. 183 According to the United = States=20 Department of Health and Human Services, African Americans are = more likely=20 to suffer from hypertension and are less likely to receive = medication than=20 the general population with hypertension. 184 Some research has = suggested a=20 correlation between hypertension and susceptibility to adverse = exposures.=20 185 For example, hypertension = may=20 contribute to the development of kidney disease. Since the kidney = is=20 responsible for filtering toxic substances from the blood,=20 hypertension-induced kidney disease may reduce the body's ability = to react=20 to toxic exposures. 186

     = In=20 addition to hypertension, other diseases that occur more = frequently in=20 minority communities can cause an increase in susceptibility to = various=20 toxicants. Diabetes, for example, reduces the body's capacity to = detoxify=20 organic solvents because of metabolic impairments resulting from = decreased=20 cardiovascular functioning. 187 While this vulnerability = should be=20 accounted for in risk assessment in general, it is necessary to = note that=20 diabetes occurs more frequently among Native Americans, African = Americans,=20 and Hispanics. 188 Similarly, liver disease, = which=20 increases *507 susceptibility to pesticides, = poly-chlorinated=20 biphenyls, metals, and hydrocarbons, occurs more frequently in = minority=20 communities. 189 Tuberculosis, which = adversely=20 affects the lung's ability to eliminate particulates, is more = common among=20 Asians, African Americans, and Hispanics. 190 Finally, asthma increases = the=20 body's susceptibility to pollutants and is more commonly found and = less=20 often treated among African Americans. 191

3. Social=20 Inequalities

     Circumstances of = social=20 inequality can adversely affect susceptibility to environmental = health=20 hazards. One consequence of both overt and institutional racism, = for=20 example, has been its independent adverse health impact which in = turn=20 affects susceptibility to environmental pollutants. 192 A lack of political power = and=20 access to information may have the most profound effect on = susceptibility=20 because with it comes an inability to effect change and thereby = lessen=20 exposure to harmful substances.

     = Low-income=20 populations generally have less access to consistent and = preventative=20 health care, including adequate child care. 193 Race in particular may = also affect=20 health care in that some studies have shown that minorities = frequently=20 receive a lower quality of care. 194

*508

 =    =20 Low income is also correlated with poor nutrition, including = insufficient=20 intake of proteins and vitamins. 195 As Edward Calabrese has=20 demonstrated, nutritional deficiencies and susceptibilities to=20 environmental pollutants have a distinct relationship. 196 For instance, vitamin C = deficiency=20 is associated with increased vulnerability to lead, benzene, = insecticides,=20 carbon monoxide, ozone, and sulfates. 197 Similarly, protein = deficiency is=20 associated with increased vulnerability to insecticides and = industrial=20 solvents; calcium deficiency is associated with increased = susceptibility=20 to lead; and iron deficiency is associated with an increased = sensitivity=20 to hydrocarbon carcinogens, lead, and manganese. 198

     = Finally, inadequate education and poor English skills among some = minority=20 populations undermine the communication necessary to facilitate=20 environmental protection. 199 Lack of information = regarding risky=20 substances, safety precautions, and regulatory safeguards is = likely to=20 affect both susceptibility and exposure to harmful=20 substances.

4. Lifestyle=20 Factors

     In addition to genetic = characteristics, disease frequencies, and social inequalities, = lifestyle=20 factors in some circumstances may adversely affect minority = populations.=20 For example, because the birth rate among many minority groups is = higher=20 than that of the general population, 200 minority populations have = larger=20 percentages of children and pregnant women. Therefore, because = pregnant=20 women, children, infants, and fetuses are more susceptible = *509 to=20 adverse health effects from pollutants than are members of the = remainder=20 of the population, 201 exposure to pollutants = will=20 disproportionately affect minority=20 communities.

     Finally, studies = indicate=20 that minority and low-income populations, regularly targeted by = alcohol=20 and tobacco advertising campaigns, 202 use these products with = greater=20 frequency than does the general population. 203 Such substance abuses are = associated with increased susceptibility to toxic chemicals by = creating=20 impaired respiratory, cardiovascular, and metabolic processes. 204

     = This Part has demonstrated some of the possible informational = distortions=20 or biases within risk assessment as they relate to susceptibility. = Individuals may differ in vulnerability to chemical substances as = a result=20 of genetic characteristics, disease frequencies, societal = inequalities,=20 and lifestyle patterns. To the degree that these determinants are=20 correlated to socioeconomic, ethnic and racial categories, poor = people and=20 communities of color will receive less environmental protection = than is=20 necessary to adequately safeguard health.

III. Possible=20 Regulatory Responses to the Environmental Justice Critique of Risk = Assessment


     The = environmental=20 justice critique of risk assessment is a multi-faceted attack = which=20 alleges, inter alia, that the risk assessment process postpones = regulatory=20 or remedial action in the face of a real health risk to a = community and=20 that risk assessment fails to incorporate community perceptions of = risk.=20 205 While a full response = *510=20 to these objections is critical in order to answer the breadth of = the=20 environmental justice critique of risk assessment, this Student = Article=20 has focused mostly on the informational biases that result from = the=20 practice of risk assessment.

     Part = II=20 catalogued the theoretical foundations supporting the claim that = the=20 practice of risk assessment contains an informational bias that=20 systematically discriminates against poor communities and = communities of=20 color. As detailed above, this bias results from a failure to = incorporate=20 information regarding exposure to multiple chemicals, chemical = mixtures,=20 above-average exposures, as well as information regarding = increased=20 susceptibility to chemicals.

A. Abandoning Risk=20 Assessment

     Before discussing = solutions=20 to such informational biases, however, it should be noted that = many view=20 risk assessment as fundamentally flawed and advocate for its = abandonment.=20 206 Such an abandonment of = risk-based=20 regulation, however, may be both impractical and=20 undesirable.

     First, risk = assessment is=20 required or its use implied in an increasing number of = environmental and=20 health statutes. 207 Second, even where risk = assessment=20 is not required, the practice has grown to become the dominant = method of=20 understanding the health effects of chemical exposures. 208 Third, risk assessment = offers at=20 least the theoretical possibility of regulating efficiently and=20 objectively. 209

*511

 =    =20 Finally, the components of risk assessment -- in particular = exposure=20 assessment and dose/response assessment -- are tools well suited = to=20 environmental justice concerns. In theory, they allow for the = input of any=20 degree of specificity of data regarding exposure levels and=20 susceptibilities. 210 Therefore, at least with = respect to=20 the informational biases discussed in Part II, an environmental = justice=20 critique does not necessarily require an abandonment of regulatory = risk=20 assessments. 211

B. Correcting = the=20 Informational Bias

     In order to = correct=20 the problems discussed above and to create a more equitable risk=20 assessment process, the following methodological and regulatory = efforts=20 should be adopted: (1) more effective and exhaustive gathering of = relevant=20 demographic data; (2) reduction of generalized exposure and = generalized=20 susceptibility estimates; (3) public participation in the various = stages=20 of risk assessment decisionmaking; (4) implementation of various=20 mechanisms to impose the burden of demonstrating safety on the = private=20 sector; and (5) abandonment of generalized risk assessments in = determining=20 cleanup standards at specific sites.

1. Information=20 Gathering

     Through the use of=20 generalized exposure assessments and generalized susceptibility=20 assumptions, risk assessment is largely an effort aimed at = estimation and=20 efficiency. By assuming that the population is largely = homogeneous, the=20 practice of risk assessment obscures correlations between = particular=20 groups and the high end of both the exposure and susceptibility=20 generalizations. Thus, a critical first response from an = environmental=20 justice *512 standpoint is to gather necessary data to = discern the=20 extent of the correlation between racial and socioeconomic groups = and=20 above-average exposures and above-average susceptibilities. 212

     = There is a broad consensus that such information gathering is = important.=20 Environmental justice activists have long argued that data gaps = are=20 critical missing pieces in our understanding of the distributional = effects=20 of environmental protection. In 1992, for example, in a report = issued to=20 the Clinton transition team, the Lawyers' Committee for Civil = Rights Under=20 Law called for improved reporting and data reference systems to = more=20 effectively collect data "on populations most exposed to = environmental=20 contamination, including synergistic effects, multiple sources and = sensitive populations." 213

     = Similarly, in its Environmental Equity report, EPA concluded = that
[i]t may also be helpful in certain cases to present=20 exposure analysis as a range of potential exposures and to take = into=20 consideration demographic characterizations of the exposed = population,=20 such as: age, gender, ethnicity and race. Likewise, quantitative = estimates of risk probabilities should be displayed as = distributions=20 across the exposed population, considering the sensitive = population=20 groups. 214=20
     While some environmental = justice=20 activists quarrel with EPA's equivocating language, 215 recent EPA statements = nonetheless=20 demonstrate an active and rigorous effort to examine the = *513=20 possibility of multiple exposures and group susceptibilities. 216 In addition, many federal = scientists are beginning to acknowledge that "research to reduce = critical=20 uncertainties in health risk assessment . . . is at the heart of = efforts=20 to evaluate and resolve questions of environmental justice." 217

     = While these examples point to an increased federal effort to = gather data=20 with respect to susceptibilities and multiple pathways to = exposure, there=20 seems to be less activity with respect to understanding the = synergistic=20 effects of multiple exposures and mixtures. An environmental = justice=20 critique of risk assessment is particularly concerned with = understanding=20 the health effects of chemical interactions because of the = correlation=20 between exposure to chemical mixtures and race and = class.

2.=20 Generalized Data

     Regulatory = actions=20 based on generalized data fail to protect those whose statistical=20 characteristics place them above a certain cut-off and overprotect = those=20 who fall below a certain cut-off. There are two ways to respond to = this=20 problem. First, as statutes and regulations have often done = (though=20 without incorporating racial or economic data), 218 the government could = attempt to=20 adopt regulations using a single standard that protects the most=20 vulnerable and the most exposed. This approach of setting only = a=20 *514 single stringent standard offers the advantage of = protecting the=20 most highly exposed and susceptible. Also, the use of a single = universal=20 standard is relatively easy to administer. One difficulty with = such an=20 approach is that in many cases it leads to regulation that is too=20 stringent and therefore inefficient. The rationale for adopting = such an=20 approach, then, is that it is better to overprotect many even at = an=20 appreciable cost to society in order to ensure that everybody is = minimally=20 protected.

     Second, the government = could=20 reduce or eliminate the use of generalized data and regulate = according to=20 relative risk. That is, standards for a particular substance would = differ=20 based upon the population in a given geographical area. While this = proposal would require more stringent standards in areas more = highly=20 exposed or highly vulnerable, it would require less stringent = standards in=20 areas where exposure is limited or where risks are de minimis. = Emission=20 standards, cleanup standards, and risk comparisons, then, would = differ=20 depending upon the particular characteristics of the exposed=20 population.

     There are advantages = and=20 disadvantages to reducing the use of generalized data. On one = hand, a more=20 refined risk assessment will respond to the informational biases = discussed=20 above while also achieving more efficient regulation. By avoiding=20 generalized data, we avoid Justice Breyer's concern with = overprotection in=20 situations in which protection is unreasonable. 219

     = On=20 the other hand, because "the cost of estimating exposure is = directly=20 proportional to the certainty of the estimate," 220 the more refined the risk = quantification, the more expensive the process of generating it. = One can=20 expect that increased certainty with respect to susceptibilities = will also=20 involve great expense.

     Independent = of=20 cost, relative risk standards have other problems of concern to = the=20 environmental justice movement. Because of the possibility that = lower=20 standards will be used in situations in which human risk and = exposure are=20 minimal, the decreased use of generalized data creates a possible = conflict=20 of interest between traditional environmentalists and the = environmental=20 justice community. Traditional environmentalists, *515 = concerned=20 with the protection of natural resources and wildlife in addition = to human=20 health, might object to relative environmental standards because = of the=20 possibility of lower standards in areas where health risk is low = (e.g.,=20 unpopulated, "natural" areas). Environmental justice activists, on = the=20 other hand, might support relative environmental standards because = of the=20 possibility of more stringent standards in situations in which = human=20 exposure and risk are high.

     This = conflict=20 between environmentalists and environmental justice activists can = be=20 overstated, however. The possibility of lower standards is limited = by a=20 statutory regime that requires protection of the environment in = addition=20 to public health. 221 The reduced use of = generalized data=20 in risk assessments, then, would correct problems at the extremes. = Where=20 exposure and risk are extremely high, standards would be more = stringent.=20 Where exposure and risk are extremely low (despite considerations = of=20 possible multiple exposures, synergistic effects, and particular=20 susceptibilities), standards may be less stringent, but not to the = degree=20 of compromising protection of the environment. 222 Where risk is low and = certainty is=20 high, standards would respond accordingly, thereby answering those = critics=20 who argue that environmental regulations are overprotective. 223

*516

3. = Public Participation

     = Consistent and=20 effective public participation in the risk assessment methodology = would=20 positively affect the informational bias within risk assessment by = improving the flow of information from the public to risk = assessors. 224 The current level of = public=20 participation in the risk assessment process is minimal. For = example, a=20 recent Government Accounting Office (GAO) report analyzing the=20 effectiveness of the Agency for Toxic Substances and Disease = Registry=20 (ATSDR) concluded that ATSDR failed to involve local communities = in the=20 process of determining risks at Superfund sites. 225

     = This lack of public participation is particularly pronounced in=20 communities of color and poor communities. For example, EPA = recently=20 acknowledged the lack of minority and poor community participation = in risk=20 assessments: "[p]oor and racial minority communities are rarely = involved=20 in Agency rulemakings and seem to be unaware . . . of their = ability to=20 petition ATSDR to perform health surveys in their communities." 226

*517=20

    Public participation in risk = assessment is=20 critical to any effort to correct the informational biases = discussed in=20 the Student Article. 227 With respect to risk = assessments=20 performed to determine cleanup standards at specific sites, = increased=20 community input regarding exposure and susceptibility assumptions = can have=20 a significant impact on risk assessment data. Residents near the = site will=20 have first-hand knowledge of particular health symptoms and=20 susceptibilities as well as possible exposure=20 pathways.

     With respect to risk = assessments=20 that are performed in order to determine emission, content, or = other types=20 of standards, the greatest potential for positive public = participation=20 lies in the hazard selection process in which agencies decide = which=20 substances require risk assessments. 228 As discussed above, a = number of=20 important risk assessments have resulted directly from petitioning = by=20 groups and individuals external to the government agency. 229 Much of the environmental = justice=20 critique of risk assessment surrounds the problems of unique = chemical=20 mixtures and synergistic effects that are not known or suspected = to be=20 dangerous. The ability to influence the selection of which hazards = to=20 assess, therefore, is of critical concern to environmental justice = activists.

     A possible mechanism = for=20 encouraging public participation is through technical assistance = grants=20 (TAGs). Currently in place in the Superfund program, 230 TAG money allows a = community to=20 hire its own scientists and technical consultants so that the = community=20 can better understand government data and can better communicate = with=20 government scientists. 231 A similar program could = be=20 implemented for highly exposed or vulnerable communities in order = for=20 those groups to either suggest possible hazards for risk = assessments or=20 perform their own preliminary assessments. While obstacles exist = within=20 the current TAG implementation, 232 *518 the TAG = program offers=20 tremendous potential for accelerating public participation in the = risk=20 assessment process.

     Two = bureaucratic=20 forces come together to undermine the possibility of effective = public=20 participation in the risk assessment process. First, regulatory = agencies=20 are more likely to create thorough, legally defensible risk = assessments=20 when a future regulatory action will follow from the assessment 233 or when the action is = likely to be=20 attacked in the courts as too lenient or strict. 234 In an effort to avoid the = court=20 overturning a particular standard, a regulatory agency will = attempt to=20 create an administrative record backed by an extensive and = detailed risk=20 assessment. 235 On the other hand, = agencies with=20 educational functions are inherently less concerned with the = explicit and=20 detailed quantification of risks. For example, while the Office of = Smoking=20 and Health has extensive data about the effects of passive = smoking, it=20 chose not to communicate its knowledge in the form of a quantified = risk=20 assessment. 236

     = Second, the anticipated use of a risk assessment discourages = agencies with=20 regulatory functions from beginning the risk assessment process = without=20 clear data indicating that a substance or mixture is hazardous. 237 On the other hand, = agencies with an=20 educational mandate are more likely to begin a risk assessment = with less=20 conclusive data indicating a hazard. As observers have *519 = noted,=20 "[o]ther agencies with different missions -- such as the National=20 Institute for Occupational Safety and Health (NIOSH), National = Cancer=20 Institute (NCI), and Office of Smoking and Health (OSH) -- do risk = assessments for education or advisory purposes and may be willing = to go=20 forward with a risk assessment based on less conclusive data." 238 This added flexibility of = nonregulatory agencies also allows those agencies to more readily = initiate=20 a study as a result of external pressures. 239

     = The=20 dynamic of these bureaucratic forces creates a disadvantage for = poor=20 communities and communities of color. Precisely those agencies = that are=20 more willing to do experimental risk assessments and are more = likely to=20 respond to external concerns are also the agencies less concerned = with=20 performing thorough risk assessments that consider varying = exposure and=20 susceptibility patterns. Furthermore, the studies that are more = likely to=20 be responsive to petitions from the public are less likely to = result in=20 regulatory action. A revised mandate should allow for regulatory = agencies,=20 in addition to educational agencies, to set aside resources for=20 experimental risk assessments. As a result, the risk assessments = that=20 respond to public pressure will be sufficiently rigorous to result = in=20 regulatory action when necessary. 240

4. Burden of=20 Proof

     Improved risk = assessments to=20 protect those more exposed and more vulnerable will incur great = costs. 241 Regulatory bodies are = already=20 reluctant to engage in an extensive risk assessment for each = possible=20 chemical combination because of the lack of adaptability to other=20 situations. One method of spreading the cost and encouraging risk=20 assessments that overcome the informational bias is to require the = private=20 sector to meet safety standards *520 prior to being allowed = to=20 release, store, or use substances. 242

     = Such a model has been applied in other environmental regulatory = contexts.=20 For example, a company wishing to comply with the Clean Air Act in = an area=20 that as a whole, is not meeting applicable air quality standards = is=20 required to offset its proposed emission by a reduction in the = same area.=20 243 Similarly, a company = wishing to=20 release a substance through any media in an area that is exposed = to=20 another substance (or several other substances) should be required = to show=20 that its release will not be harmful. Such a burden would require = a=20 showing that neither additivity nor synergism is likely to produce = an=20 adverse health effect in the exposed area. Similarly, such a = burden of=20 proof would require a susceptibility analysis that incorporates=20 demographic variables.

     While the = cost of=20 such a burden may be great, it could be mitigated by a public = commitment=20 to develop appropriate tools with which the private commercial = sector=20 could perform extensive site-specific assessments. In exchange for = providing sophisticated, reliable models of chemical interaction, = for=20 example, the government would be relieved from having to perform = actual=20 assessments on millions of possible combinations.

5. = Cleanup=20 Standards

     Perhaps the most = egregious=20 and concrete error of risk assessment involves the setting of = cleanup=20 standards for Superfund sites. To the degree that Superfund sites = -- and=20 state counterparts -- are waste baskets for hundreds of different=20 substances, applications of risk assessments from single substance = studies=20 are increasingly arbitrary. Many community activists and = representatives=20 provide persuasive accounts of residents who are mysteriously = *521=20 ill notwithstanding government assurances of absolute safety. 244

     = Certainly, the use of risk assessments in such situations must be = refined=20 to include actual medical examinations of those residents exposed = to the=20 site in addition to studies of the actual combinations present. = Also, any=20 necessary risk assessments should be expedited and should not = occur until=20 it is clear that nearby residents are not in immediate=20 danger.

     Refining risk assessments = for=20 cleanup sites should be a noncontroversial proposal. First, nearby = residents -- in addition to receiving immediate medical attention = and=20 evacuation if necessary -- will benefit from site- specific = studies that=20 account for the particular mixtures to which they are exposed. 245 At the same time, a more=20 site-specific risk assessment at Superfund sites would help = alleviate many=20 of the problems associated with over- cleaning and unreasonable = standards=20 in cases where there is no threat to human health. 246


Conclusion
=
    =20 Risk assessment is an increasingly integral aspect of government = safety=20 regulation. The government's health research efforts -- as = manifest=20 through risk assessments -- may systematically cause poor and = minority=20 communities to receive less than adequate environmental protection = and=20 thus experience disproportionately adverse health effects. The = legitimacy=20 of a public program that relies on risk assessments, therefore, is = undermined *522 to the degree that the underlying risk = assessments=20 distort or hide certain risks. This Student Article articulates = the=20 possible ways in which distortions occur and places them within = the=20 context of the prevailing risk assessment=20 methodologies.

     Risk assessment = methodology=20 currently incorporates numerous informational biases that may=20 disproportionately affect poor communities and communities of = color.=20 Specifically, risk assessments generally fail to observe those = adverse=20 health effects that result from above-average exposure, from = exposure to=20 multiple chemicals, and from the interactions of chemicals. = Similarly,=20 risk assessments generally fail to observe susceptibility = differences as a=20 function of race or income. Genetic differences, disease patterns, = social=20 inequalities, and cultural and lifestyle factors all, in certain=20 circumstances, increase the body's susceptibility to chemical=20 substances.

     In order to protect = those most=20 exposed and those most susceptible, risk assessment must account = for these=20 factors. Permit requirements and site- specific cleanup standards = should=20 be examined to improve the efficacy and equity of our regulatory = efforts.=20 Also, greater efforts at collecting information, incorporating = refined=20 exposure and susceptibility data, and increasing public = participation must=20 occur if environmental regulations based on scientific risk = assessments=20 are to approach the goal of equal protection.=20


------=_NextPart_001_0003_01C619F5.950D5A40-- ------=_NextPart_000_0000_01C619F5.950D5A40 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Content-Location: http://www.law.nyu.edu/journals/envtllaw/issues/vol3/2/3nyuelj469n.html NYU ELJ NOTES
=86 B.A., = 1991, New=20 College, University of South Florida; J.D., 1995, New York = University=20 School of Law (expected). The author was Co-Editor-in-Chief for = the N.Y.U.=20 Environmental Law Journal during the 1994 - 95 academic year. The = author=20 gratefully acknowledges the assistance of Vicki Been, Deeohn = Ferris, Adam=20 Finkel, Nancy Marks, Vernice Miller, and Richard Revesz.

1 "The keystone of this = quest for=20 justice is equal protection, not equal pollution." Deeohn Ferris, = A=20 Challenge to EPA, EPA J., Mar. 1992, at 28, 28. See generally = Michel=20 Gelobter, Toward a Model of "Environmental Discrimination," in = Race and=20 the Incidence of Environmental Hazards: A Time for Discourse 64, = 73 - 80=20 (Bunyan Bryant & Paul Mohai eds., 1992) (providing a model for = understanding environmental discrimination as the interaction of = formal=20 and informal government decisions and external structural = processes);=20 Rachel D. Godsil, Comment, Remedying Environmental Racism, 90 = Mich. L. Rev=20 . 394 (1991) (proposing an environmental "disparate impact" = statute=20 modeled after the Civil Rights Act of 1990); Richard J. Lazarus, = Pursuing=20 "Environmental Justice": The Distributional Effects of = Environmental=20 Protection, 87 Nw. U. L. Rev. 787 (1993) (showing how the = distribution of=20 environmental benefits and burdens accrued through regulation = adversely=20 affect minority and poor communities and suggesting the use of = civil=20 rights law as a possible remedy); Peter L. Reich, Greening the = Ghetto: A=20 Theory of Environmental Race Discrimination, 41 Kan. L. Rev. 271 = (1992)=20 (arguing that state rather than federal equality jurisprudence = offers an=20 accessible avenue toward environmental justice).

2 See, e.g., Robert D. = Bullard,=20 Dumping in Dixie: Race, Class and Environmental Quality 42- 43 = (1992)=20 (noting that in Houston, Texas, six of the city's eight waste=20 incinerators, five of the city's privately owned landfills, and = all five=20 of the city's publicly owned landfills are in areas where the = majority of=20 the residents are African American); Commission for Racial = Justice, United=20 Church of Christ, Toxic Wastes and Race in the United States: A = National=20 Report on the Racial and Socio-Economic Characteristics of = Communities=20 with Hazardous Waste Sites xiv (1987) [hereinafter Toxic Wastes = and Race ]=20 (finding that "[t]hree out of every five Black and Hispanic = Americans=20 lived in communities with uncontrolled toxic waste sites"); United = States=20 General Accounting Office, Siting of Hazardous Waste Landfills and = Their=20 Correlation with Racial and Economic Status of Surrounding = Communities=20 (1993) (showing that three out of four commercial hazardous waste = sites in=20 eight southern states were located in majority black communities); = Paul=20 Mohai & Bunyan Bryant, Environmental Racism: Reviewing the = Evidence,=20 in Race and the Incidence of Environmental Hazards: A Time for = Discourse,=20 supra note 1, at 163, 166 - 67 (reviewing eighteen studies and = concluding=20 that race is a factor independent of class in the disproportionate = siting=20 of hazardous facilities).
While there is broad consensus = regarding the=20 disproportionate distribution of environmentally hazardous or = undesirable=20 facilities, there is less agreement regarding the cause of such a=20 distribution. See, e.g., Vicki Been, Locally Undesirable Land Uses = in=20 Minority Neighborhoods: Disproportionate Siting or Market = Dynamics?, 103=20 Yale L.J. 1383 (1994) (presenting data suggesting that market = dynamics=20 such as property values may cause some of the observed = distributional=20 problem).

3 Robert D. Bullard, The=20 Environmental Justice Framework: A Strategy for Addressing Unequal = Protection (1992) (unpublished document), quoted in Deeohn Ferris, = Testimony Before the Subcomm. on Superfund, Recycling, and Solid = Waste=20 Management of the Senate Comm. on Environment and Public Works 4 = (July 28,=20 1993) (unpublished document, on file with the N.Y.U. Environmental = Law=20 Journal).

4 Marianne Lavelle & = Marcia=20 Coyle, Unequal Protection: The Racial Divide in Environmental Law, = Nat'l=20 L.J. , Sept. 21, 1992, at S1. To be sure, allegations of unequal=20 environmental enforcement are controversial. See, e.g., Skreekant = Gupta,=20 George L. Van Houtven & Maureen L. Cropper, Do Benefits and = Costs=20 Matter in Environmental Regulation? An Analysis of EPA Decisions = Under=20 Superfund, in Analyzing Superfund (Richard Revesz & Richard = Stewart=20 eds., forthcoming 1995) (finding no disparate enforcement at = Superfund=20 sites and criticizing Lavelle and Coyle for not controlling for = variables=20 other than race and income).

5 Lavelle & Coyle, = supra note 4,=20 at S4. According to the study, communities of color as defined by = the=20 study will wait up to four years longer than white communities to = get a=20 Superfund site cleaned up.

6 "Congress said that = wherever=20 possible, hazardous waste sites should be cleaned up by destroying = or=20 treating the dangerous material rather than just containing or = isolating=20 it. At sites where a cleanup decision has been made, those located = in=20 white communities are more likely to get the preferred treatment = than=20 sites in minority areas." Lavelle & Coyle, supra note 4, at = S6.=20 Specifically, in white areas, according to the National Law = Journal, 22%=20 more sites are treated than contained, while in minority areas, 7% = more=20 sites are contained than treated. Lavelle & Coyle, supra note = 4, at=20 S6.

7 Fines under the Resource = Conservation and Recovery Act (RCRA), 42 U.S.C. = =A7=A7=20 6901-6992k=20 (1988 & Supp. IV 1992), averaged $335,566 in "white areas" and = $55,318=20 in "minority areas" according to the National Law Journal. Lavelle = &=20 Coyle, supra note 4, at S2.

8 See, e.g., Southwest = Network for=20 Environmental and Economic Justice, Comments to the Environmental=20 Protection Agency Environmental Equity Report, in 2 U.S. = Environmental=20 Protection Agency, Environmental Equity: Reducing Risk For All = Communities=20 90, 93 (1992) [hereinafter SNEEJ Comments].

9 See generally William D. = Ruckelshaus, Risk, Science, and Democracy, in Readings in Risk 105 = (Theodore S. Glickman & Michael Gough eds., 1990) (stating = that risk=20 was not a central element of early environmental regulation in = this=20 country and that the initial focus at EPA in the early 1970s was = the=20 reduction of pollution). William Ruckelshaus, administrator of the = EPA=20 from 1970 to 1973, attributes EPA's increasing risk focus in the = 1970s and=20 1980s to public concern over cancer-causing substances, such as = PCBs and=20 asbestos. Id. at 109. Once public concern shifted to the potential = cancer=20 risks of exposure to any of a large number of substances, EPA = became more=20 interested in trying to identify and quantify those risks: "the = problem of=20 uncertainty was moved from the periphery to the center." Id.; see = also=20 Task Force on Health Risk Assessment, U.S. Department of Health = and Human=20 Services , Determining Risks to Health: Federal Policy and = Practice 304=20 (1986) [hereinafter DHHS] ("The scope and importance of DHHS = activities to=20 determine health risks has grown beyond the point where risk = assessment=20 can be treated as a narrow discipline. It is now a major = departmental=20 activity, and the recent trend of growth is likely to=20 continue.").

10 It should be noted that = an=20 environmental justice critique of risk assessment does not = specifically=20 allege that minorities and poor people receive less environmental = benefit=20 than others but that these groups receive less than adequate = benefit to=20 protect health. In fact, some studies have indicated that = minorities and=20 the poor are often the greatest beneficiaries of environmental = programs.=20 See e.g., Leonard P. Gianessi, Henry M. Peskin & Edward Wolff, = The=20 Distributional Effects of Uniform Air Pollution Policy in the = United=20 States, 93 Q.J. Econ. 381, 293 -300 (1979) (showing that racial = minorities=20 are beneficiaries of the Clean Air Act). However, this does not=20 necessarily dilute the claim that the benefit received in those=20 communities is less than adequate whereas the benefit received in = wealthy=20 or white communities is adequate to protect human health. For = reasons=20 supporting this claim see infra part II.

11 Specifically, the = argument that=20 distorted risk assessments preclude the possibility of equal = protection=20 from pollution rests on the claim that government regulation must = result=20 in similar health status across groups, to the extent that health = status=20 is affected by substances that are regulated. This is different = than the=20 claim that government efforts must be equal across groups, or that = government efforts must result in an equal decrease in pollution = across=20 groups. This Student Article does not argue that government = efforts differ=20 across groups; rather, it suggests that the same government effort = may be=20 adequate for one group and inadequate for another. For a = discussion of the=20 meaning of fairness in the context of siting undesirable = facilities, see=20 Vicki Been, What's Fairness Got To Do With It? Environmental = Justice and=20 the Siting of Locally Undesirable Land Uses, 78 Cornell L. Rev. = 1001=20 (1993) (presenting the philosophical and pragmatic challenges = posed by=20 seven different meanings of environmental justice).

12 See e.g., 2 U.S. = Environmental=20 Protection Agency, Environmental Equity: Reducing Risk For All = Communities=20 =A7 5.0 (1992) [hereinafter EPA Equity Report ].

13 The Food and Drug = Administration=20 (FDA), for example, is directly involved in risk assessment in its = efforts=20 to ensure that our food supply is safe. The mandated risk = limitation for=20 food additives is the highly stringent Delaney Clause, which = states that=20 "no additive shall be deemed to be safe if it is found to induce = cancer in=20 man or animal." Food, Drug, and Cosmetic Act of 1938, 21 U.S.C. = =A7=A7 301-393, = 348(c)(3)(A) = (1988 & Supp. IV 1992); see also Federal Insecticide, = Fungicide, and=20 Rodenticide Act, 7=20 U.S.C. =A7=A7 136-136y, = 136bb = & 136(a)(c) = (1988=20 & Supp. IV 1992) (requiring that pesticides not cause "any=20 unreasonable risk to man or the environment, taking into account = the=20 economic, social, and environmental costs and benefits of the use = of any=20 pesticide").

14 Clean Air Act, 42 U.S.C. = =A7=A7=20 7401-7671q=20 (1988 & Supp. IV 1992).

15 Clean Water Act, 33 U.S.C. = =A7=A7=20 1251-1387=20 (1988 & Supp. IV 1992); Safe Drinking Water Act, 42 U.S.C. = =A7=A7 300f=20 to 300j-26=20 (1988 & Supp. IV 1992).

16 Solid Waste Disposal = Act, 42 U.S.C. = =A7=A7=20 6901-6992k=20 (1988 & Supp. IV 1992), as amended by Resource Conservation = and=20 Recovery Act, 42=20 U.S.C. =A7=A7 6901-6992k = (1988=20 & Supp. IV 1992); Comprehensive Environmental Response, = Compensation,=20 and Liability Act, 42 U.S.C. = =A7=A7=20 9601-9675=20 (1988 & Supp. IV 1992).

17 The Occupational Safety = and=20 Health Act of 1970 established the National Institute for = Occupational=20 Safety and Health (NIOSH) within the Center for Disease Control = (CDC). 29 U.S.C. = =A7=A7 651-678, = 671 = (1988 &=20 Supp. IV 1992). NIOSH and the Secretary of Labor employ risk = assessment=20 techniques in order to comply with their obligation to establish = and=20 promulgate safety and health standards for employees. 29 U.S.C. = =A7=A7 654-655 = (1988 &=20 Supp. IV 1992).

18 Indoor pollution is = regulated=20 through the Clean Air Act, 42 U.S.C. = =A7=A7=20 7401-7671q=20 (1988 & Supp. IV 1992), the Safe Drinking Water Act, 42 U.S.C. = =A7=A7 300f=20 to 300j-26=20 (1988 & Supp. IV 1992), and the Toxic Substances Control Act, = 15 U.S.C. = =A7=A7=20 2601-2692=20 (1988 & Supp. IV 1992).

19 Toxic Substances = Control Act, 15 U.S.C. = =A7=A7=20 2601-2692=20 (1988 & Supp. IV 1992). "[A]mong the many chemical substances = and=20 mixtures which are constantly being developed and produced, there = are some=20 whose manufacture, processing, distribution in commerce, use, or = disposal=20 may present an unreasonable risk of injury to health or the = environment."=20 15 = U.S.C. =A7=20 2601(a)(2) (1988 & Supp. IV 1992); Consumer Product Safety = Act of=20 1972, 15 U.S.C.=20 =A7=A7 2051-2084=20 (1988 & Supp. IV 1992).

20 For general discussions = of the=20 role of risk assessment in the regulatory process see Assessing = Risks to=20 Health: Methodological Approaches (John C. Bailar III et al. eds., = 1993)=20 [hereinafter Assessing Risks to Health ]; Stephen Breyer , = Breaking the=20 Vicious Circle: Toward Effective Risk Regulation (1993); Committee = on the=20 Institutional Means for Assessment of Risks to Public Health, = National=20 Research Council, Risk Assessment in the Federal Government: = Managing the=20 Process (1983); DHHS, supra note 9; Ronald E. Gots , Toxic Risks: = Science,=20 Regulation, and Perception (1993); Quantitative Risk Assessment in = Regulation (Lester B. Lave ed., 1982); Readings in Risk (Theodore = S.=20 Glickman & Michael Gough eds., 1990); Jeffrey S. Lubbers, Risk = Regulation at the Federal Level: Administrative Procedure = Constraints and=20 Opportunities, 1 Risk 43 (1990); Junius C. McElveen, Jr., Risk = Assessment=20 in the Federal Government: Trying to Understand the Process, 5 = Tul. Envtl.=20 L.J. 45 (1991); Alon Rosenthal et al., Legislating Acceptable = Cancer Risk=20 from Exposure to Toxic Chemicals, 19 Ecology L.Q. 269 = (1992).

21 Pharmacokinetics is the = science=20 of the "absorption, distribution within the body, metabolism, and=20 excretion of toxic chemicals." Rosenthal et al., supra note 20, at = 283=20 n.74.

22 DHHS, supra note 9, at=20 8.

23 EPA Equity Report , = supra note=20 12, at 30.

24 Barbara L. Berney, = Ethylene=20 Dibromide, in Assessing Risks to Health , supra note 20, at 21,=20 42.

25 See, e.g., Ethyl Corp. = v. EPA,=20 541 F.2d 1, 20 -24 (D.C. Cir. 1976) (en banc) (upholding = regulation of=20 lead additives in gasoline on basis of risk assessment); Reserve = Mining=20 Co. v. EPA, 514 F.2d 492, 507-20 (8th Cir. 1975) (en banc) = (upholding=20 regulation of asbestifom fibers on basis of EPA's informal risk = assessment=20 which considered toxicological and epidemiological = evidence).

26 Industrial Union Dep't, = AFL-CIO=20 v. American Petroleum Inst., 448 U.S. 607 (1980). Other factors = that have=20 led to the growth of risk assessment include President Reagan's = Executive=20 Order No. 12,291, 46 Fed. Reg. 13,193 (1981), which required = cost-benefit=20 justifications for major regulations. See John S. Applegate, The = Perils of=20 Unreasonable Risk: Information, Regulatory Policy, and Toxic = Substances=20 Control, 91 Colum. L. Rev . 261, 281- 82 (1991).

27 Thomas Bartman, = Regulating=20 Benzene, in Quantitative Risk Assessment in Regulation , supra = note 20, at=20 99, 99.

28 Id. at 101- = 02.

29 42 Fed. Reg. 27,452, = 27,455=20 (1977). The proposed standard incorporated a one part per million = of=20 benzene concentration in air over an eight hour shift. The final = standard=20 was issued on February 10, 1978. See 29 C.F.R. =A7 1910.1028=20 (1993).

30 Industrial Union Dep't, = 448 U.S.=20 607.

31 Id. at 630.

32 Id. at = 631-38.

33 29 U.S.C. = =A7=20 652(8) (1988).

34 Industrial Union Dep't, = 448 U.S.=20 at 642 (stating that unsafe employment occurs when "significant = risks are=20 present and can be eliminated or lessened by a change in=20 practices").

35 Id. at 653.

36 Id. at 655.

37 Id. at 656.

38 Jack Needleman, The = Role of Risk=20 Assessment in Agency Activities, in Assessing Risks to Health , = supra note=20 20, at 9, 16.

39 See Two Faces of Risk, = EPA J.,=20 Jan./Feb./Mar. 1993, at 19, 19.

40 This is the function of = risk=20 assessment that was at issue in the benzene case. See supra notes = 26 -38=20 and accompanying text.

41 EPA's Superfund program = administers the Comprehensive Environmental Response, = Compensation, &=20 Liability Act (CERLCA), 42 U.S.C. = =A7=A7=20 9601-9675=20 (1988 & Supp. IV 1992).

42 42 U.S.C. = =A7=20 9605(c)(1) (1988). See generally McElveen, supra note 20, at = 73 - 99=20 (describing the history and mechanics of the HRS).

43 42 U.S.C. = =A7=20 9605(c)(1) (1988). Comparative risk assessment is also = employed in=20 EPA's regulation of drinking water, pesticides, surface water = pollution,=20 air pollution, waste disposal, and toxic chemicals. See Donald T.=20 Hornstein, Reclaiming Environmental Law: A Normative Critique of=20 Comparative Risk Analysis, 92 Colum. L. Rev . 562, 568 nn.20 -25=20 (1992).

44 "Some assessments are=20 retrospective, focusing on injury after the fact -- for example, = the kind=20 and extent of risks at a particular Superfund site. Others seek to = predict=20 possible future harm to human health or the environment -- for = example,=20 the risks expected if a newly developed pesticide is approved for = use on=20 food crops." Dorothy E. Patton, The ABCs of Risk Assessment, EPA = J.,=20 Jan./Feb./Mar. 1993, at 10, 10.

45 See generally Gene A. = Lucero,=20 Superfund: Remedy Selection After SARA, 377 PLI/Lit 167 (1989) = (stating=20 that if no pre-existing applicable standards exist, then EPA must = perform=20 a risk assessment to determine whether contamination levels = threaten=20 public health).

46 "Throughout 1994, = political=20 observers have declared that no environmental bill would pass = Congress=20 without a significant risk policy amendment attached to it. Now, = as the=20 103rd Congress winds down, it is clear that risk proposals are, = indeed,=20 clogging the legislative works." Chances for Risk Legislation this = Year=20 Appear Slim, Inside EPA's Risk Policy Report , Sept. 16, 1994, at = 15, 15=20 (surveying the various legislative attempts at improving the = quality of=20 risk assessments and expanding the role of risk assessment within=20 environmental and health regulation).

47 According to Carol = Browner,=20 Administrator of EPA, "[r]isk assessment is an important and = reasonable=20 tool, but requiring EPA to use risk assessment in every instance = is not=20 reasonable." Gary Lee, Analyzing Risk Assessment at EPA: Some See = It as=20 Management Tool; Others Call It Unreasonable, Wash. Post , Mar. 8, = 1994,=20 at A17. Browner also argued that "in many instances, risk = assessment has=20 become a code word for those who want to weaken our efforts to = protect=20 public health and the environment." Id.

48 Deeohn Ferris, a = leading attorney=20 in the environmental justice movement and Executive Director of = the=20 Washington Office on Environmental Justice, has put forth this = argument=20 stating that "[w]hile we may not, in the near term, be able to = establish=20 immediate cause and effect, we certainly ought to understand that = we=20 should not wait until we have a body count to regulate those = effects that=20 are so harmful." James Lawlor, From Clinton Order to 1964's Title = 6, "E"=20 Justice Advances on the Agenda, Air Water Pollution Report, Feb. = 28, 1994,=20 available in LEXIS, Envirn Library, Curnws File.

49 For example, Senator = Max Baucus=20 (D-Mont.) recently responded to a proposal by Representative John = Mica=20 (R-Fla.) to mandate EPA to perform a risk assessment for all = regulations=20 by arguing that "EPA worked for years to develop a health = risk-based=20 approach to air pollution. It proved difficult. Regulating by risk = assessment alone, in this and I believe in most cases, is too = inflexible=20 and open to scientific question for efficient enactment and = enforcement."=20 Lee, supra note 47, at A17.

50 Contract With America = 125 -32 (Ed=20 Gillespie & Bob Schellhas eds., 1994) (proposing the "Job = Creation and=20 Wage Enhancement Act" which would require risk assessments for all = regulations in an effort to correct the perceived current = situation where=20 federal regulators, "attack whatever health risk has caught the = public's=20 attention"); Senator Daniel P. Moynihan, A Legislative Proposal, = EPA J. ,=20 Jan./Feb./Mar. 1993, at 46, 46 ("[R]isk ranking and cost-benefit = analysis=20 are valuable tools for making environmental decisions."); Johnston = Promises to Tackle Risk Issue Head-on in 104th Congress, Inside = EPA's Risk=20 Policy Report , Nov. 18, 1994, at 24, 24 -25 (reporting Senator = Johnston's=20 (D-La.) intention to displace "superstition" with science by = introducing=20 legislation requiring EPA to undertake risk assessments for all = major=20 regulations).

51 See generally = Hornstein, supra=20 note 43 (identifying the normative problems that arise when = environmental=20 policy is made by comparative risk analysis).

52 See supra part = I.A.2.

53 See, e.g., Robert D. = Bullard,=20 Unequal Environmental Protection: Incorporating Environmental = Justice in=20 Decision Making, in Worst Things First? The Debate Over Risk-Based = National Environmental Priorities 237 (Adam M. Finkel & = Dominic=20 Golding eds., 1995) (arguing that environmental protection is a = right);=20 see also Steven Kelman, Cost-Benefit Analysis: An Ethical = Critique, in=20 Readings in Risk , supra note 20, at 129, 129 ("In areas of = environmental,=20 safety, and health regulation, there may be many instances where a = certain=20 decision might be right even though its benefits do not outweigh = its=20 costs.").

54 Contract With America , = supra=20 note 50, at 131 (arguing that currently "Congress is never forced = to=20 ensure that the benefits of regulation ... outweigh the costs, = lost jobs,=20 and lower wages"); Moynihan, supra note 50, at 46 ("People want a = clean=20 environment, but they don't want to pay more than is necessary. We = can't=20 do everything at once. The question of priorities arises, almost=20 unbidden."); see also Wendy Cleland-Hamnett, The Role of = Comparative Risk=20 Analysis, EPA J. , Jan./Feb./Mar. 1993, at 18, 18 (describing the=20 importance of comparative risk assessment and the role of such = analysis=20 within regulatory decision making).

55 See, e.g., Breyer , = supra note=20 20, at 42-50; see infra note 57.

56 Linda Greer of the = Natural=20 Resources Defense Council states that "some in the environmental = and=20 grassroots community argue that risk assessments are inherently = flawed=20 because of statistical and scientific failings in the risk = assessment=20 process, which results in an underestimation of the real risks." = Viki=20 Reath, Environmental Community Takes Aim at Superfund Reform Plan, = Env't=20 Week, Mar. 3, 1994, available in LEXIS, Envirn Library, Curnws=20 File.

57 See, e.g., Breyer, = supra note 20,=20 at 48 ("These uncertainties, knowledge gaps, default assumptions, = guesses=20 and communication difficulties, all embodied in the technical = regulatory=20 process ... produce a system that with respect to those average = health=20 risks that come to its attention, sometimes produces ... = conservative=20 results that 'err on the safe side.' "); Hazardous Waste Cleanup = Project,=20 Exaggerating Risk: How EPA's Risk Assessments Distort the Facts at = Superfund Sites Throughout the United States (1993) (arguing that = the use=20 of assumptions instead of actual data, the use of unrealistic = worst-case=20 values in those assumptions, and the use of single values instead = of=20 numerical ranges result in cleanup standards which are overly = protective=20 and irrational); Robert J. Scheuplein, Uncertainty and the = "Flavors" of=20 Risk, EPA J. , Jan./Feb./Mar. 1993, at 16, 17 ("The recasting of = the=20 regulatory problem away from probable risk ... to worst case risk = results=20 in the under appreciation of risk-lowering factors.") (emphasis in = original).

58 Committee On The = Institutional=20 Means For Assessment of Risks To Public Health, National Research = Council,=20 Risk Assessment in the Federal Government: Managing the Process = (1983)=20 [hereinafter National Research Council ].

59 Id. at 7, 151-53. But = see Howard=20 Latin, Good Science, Bad Regulation, and Toxic Risk Assessment, 5 = Yale J.=20 on Reg . 89, 93 - 94 (1988) (arguing that the efforts by federal = agencies=20 to distinguish science from policy creates a false and problematic = impression that risk assessment is an objective = science).

60 National Research = Council , supra=20 note 58, at 151-53.

61 National Research = Council , supra=20 note 58, at 162-71.

62 EPA has published = guidelines for=20 many stages and types of risk assessment. See, e.g., Guidelines = for=20 Carcinogen Risk Assessment, 51 Fed. Reg. 33,992 (1986); Guidelines = for=20 Estimating Exposures, 51 Fed. Reg. 34,042 (1986); Guidelines for = the=20 Health Assessment of Suspect Developmental Toxicants, 51 Fed. Reg. = 34,028=20 (1986); Guidelines for the Health Risk Assessment of Chemical = Mixtures, 51=20 Fed. Reg. 34,014 (1986); Guidelines for Mutagenicity Risk = Assessment, 51=20 Fed. Reg. 34,006 (1986); Office of Emergency and Remedial Response = ,=20 EPA/540/1- 89/002, Environmental Protection Agency, Risk = Assessment=20 Guidance for Superfund: Volume I - Human Health Evaluation Manual = (Part=20 A), Interim Final (1989). EPA incorporates into most of its risk=20 assessment guidelines the following description: "Risk assessment = includes=20 one or more of the following components: hazard identification, = dose-=20 response assessment, exposure assessment, and risk = characterization."=20 Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,992, = 33,993=20 (1986). For similar use of the four stages of risk assessment in = other=20 agencies, see National Institutes of Health, Human Health and the=20 Environment: Some Research Needs 32 (1992); DHHS, supra note 9, at = 8.

63 Edward J. Calabrese, = Multiple=20 Chemical Interactions 601 (1991) (hereinafter Calabrese, Multiple = Chemical=20 Interactions ); see also Diane K. Wagener et al., Equity in = Environmental=20 Health: Data Collection and Interpretation Issues, 9 Toxicology = &=20 Indus. Health 775, 783 (1993) ("Data systems that compile = information on=20 pollutant concentrations in the environment are generally focused = on=20 single, or simple, forms of pollutants; complex mixtures are not = assessed=20 due to limitation of cost and proper procedures.").

64 Single substance risk=20 assessments, for example, fail to assess the risk posed by = exposure to=20 multiple chemicals and chemical mixtures. These problems are = particularly=20 important in the context of an environmental justice critique of = risk=20 assessment. See discussion infra parts II.A.1, II.A.2.

65 For example, the = authors of a=20 recent study which examined interdisciplinary approaches to = studying risk=20 concluded that "toxicological studies of chemical mixtures are = essential=20 for understanding human risk in today's environment ... [S]ince = the Clean=20 Air Act requires regulation of individual chemicals, we designed = this=20 research program to address individual chemicals and simple = mixtures."=20 E.C. Grose et al., Interdisciplinary Approach to Assessing the = Health Risk=20 of Air Toxic Chemicals: An Overview, in Health Hazards and Risks = from=20 Exposure to Complex Mixtures and Air Toxic Chemicals 39, 47 (M.A. = Mehlman=20 ed., 1991).

66 Needleman, supra note = 38, at 9, 9=20 -10.

67 DHHS, supra note 9, at=20 10.

68 Barbara L. Berney & = Jack=20 Needleman, Setting Priorities for Risk Assessment, in Assessing = Risks to=20 Health, supra note 20, at 247, 247-50. The relative ease of = identifying=20 cancer is especially evident with respect to lead, where the = contending=20 endpoint is central nervous system and behavior effects which are=20 particularly difficult to measure. Likewise, reproductive effects = are=20 difficult to measure in large part because of the number of = different=20 endpoints such as "sperm changes, ovarian cycle changes and = menstrual=20 disorders, sexual dysfunction, spontaneous abortion, birth = defects, etc."=20 Id. at 250.

69 Ruckelshaus, supra note = 9, at=20 109.

70 For example, cancer is = thought to=20 be the most sensitive endpoint for Tris (2,3 -dibromopropyl) = phosphate,=20 ethylene dibromide, dioxin, and formaldehyde. Berney & = Needleman ,=20 supra note 68, at 249.

71 As a number of = scientists have=20 observed,
[o]nce a chemical is identified as a potential = carcinogen,=20 non-cancer studies often are not pursued, even though the = compound may=20 be a toxicant in other respects. It is conceivable that a = chemical with=20 a low cancer unit risk might be a potent teratogen, but without = a=20 multidisciplinary approach, this will never be known. =
Grose=20 et al., supra note 65, at 47. This, of course, affects all of = society but=20 will disproportionately affect those who are most exposed. Other = problems=20 of risk assessment associated with communities who are highly = exposed are=20 discussed infra part II.A.

72 Elements which EPA has=20 articulated for the hazard identification of carcinogens, for = example,=20 include information regarding physical and chemical properties, = routes and=20 patterns of exposure, structure-activity relationships, metabolic = and=20 pharmocokinetic properties, toxicological effects, animal tests, = and human=20 tests. Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. = 33,992,=20 33,994 - 95 (1986).

73 EPA has explained = specific=20 criteria for determining the weight to be given to a study and = issued=20 instructions for identifying uncertainties and weaknesses in the=20 literature. Id. at 33,996.

74 Needleman, supra note = 38, at=20 12.

75 Needleman, supra note = 38, at 12.=20 An example of a threshold hazard occurs when the suspected = substance is=20 thought to cause adverse reproductive effects.

76 Needleman, supra note = 38, at 12.=20 The most important example of a non- threshold health effect is=20 carcinogenesis.

77 Needleman, supra note = 38, at=20 13.

78 Needleman, supra note = 38, at=20 13.

79 Guidelines for = Carcinogen Risk=20 Assessment, 51 Fed. Reg. 33,992, 33,997 (1986) ("[A] number of=20 mathematical models have been developed to extrapolate from high = to low=20 dose.").

80 Id. at 33,997- 98. = There are, of=20 course, many variations of the linear model, with different = assumptions.=20 Many models incorporate both linear and nonlinear assumptions at = different=20 points. Id.; see also Needleman, supra note 38, at 13.

81 The general consensus = seems to be=20 that risk assessment assumptions, including the linear = extrapolation=20 model, overstate risk at low doses. See supra note 57. But see = Adam M.=20 Finkel, Is Risk Assessment Really Too Conservative?: Revising the=20 Revisionists, 14 Colum J. Envtl. L. 2 (1989) (presenting nine=20 misperceptions of conservative assumptions and concluding that=20 conservative assumptions do not necessarily overstate risk). The = question=20 of whether and to what degree risk assessments overstate risk is = beyond=20 the scope of this Student Article. Nevertheless, an environmental = justice=20 critique of risk assessment offers two possible observations to = this=20 debate: (1) even if risk assessment overstates risk, it may = overstate risk=20 less for certain subgroups thereby reducing the margin of safety = for those=20 groups; and (2) any effort to improve the scientific accuracy or=20 sophistication of risk assessment must not only correct = methodologies that=20 tend to exaggerate risk but also those that hide risk.

82 For a general = discussion of=20 possible distortions caused by a failure to assess sub-group=20 susceptibilities, see part II.B.

83 51 Fed. Reg. 33,992, = 33,997=20 (1986).

84 Latin, supra note 59, = at 148=20 & n.242.

85 EPA Equity Report , = supra note=20 12, at 33 -34.

86 EPA Equity Report , = supra note=20 12, at 33; see also Science Policy Council, Report of the EPA = Science=20 Policy Council on Addressing "Science and Judgment in Risk = Assessment," A=20 Report By the National Research Council, in Inside EPA's Clean Air = Report=20 , June 30, 1994, at 19, 22-23 (identifying the lack of emphasis on = subgroup susceptibility as a major area requiring risk assessment=20 improvement).

87 As defined by EPA,
[a]n exposure assessment describes the intensity, = frequency,=20 and duration of contact [between a chemical substance and the = outer=20 boundary of the human body], and often evaluates the rates at = which the=20 chemical crosses the boundary (chemical intake or uptake rates), = the=20 route by which it crosses the boundary (exposure route; e.g., = dermal,=20 oral, or respiratory), and the resulting amount ... absorbed = (internal=20 dose).
Guidelines for Exposure Assessment, 57 Fed. = Reg.=20 22,888, 22,891 (1992).

88 57 Fed. Reg. 22,888=20 (1992).

89 See infra part = II.A.

90 EPA Equity Report , = supra note=20 12, at 34.

91 See, e.g., Ann Misch, = Assessing=20 Environmental Health Risks, in State of the World: 1994, at 117, = 135=20 (Lester R. Brown et al. eds., 1994) ("Most risk assessments ignore = the=20 fact that exposure to toxic chemicals is unequal and rely instead = on=20 estimates of 'average' exposure levels."). Nonetheless, many = observers=20 criticize risk assessment for incorporating worst-case exposure = scenarios.=20 See e.g., Bernard D. Goldstein, The Maximally Exposed Individual: = An=20 Inappropriate Basis for Public Health Decisionmaking, Envtl. F. ,=20 Nov.-Dec. 1989, at 13 (arguing that the risk to the aggregate = population,=20 not the most exposed individual within that population, should be = the=20 basis for regulatory decisionmaking); supra note 57. For an = explanation of=20 the value of an environmental justice critique of risk assessment = despite=20 conservative assumptions, see supra note 81.

92 "The EPA has not = employed=20 multi-path/multi-source exposure assessment routinely in all = programs."=20 Science Policy Council, supra note 86, at 22.

93 Science Policy Council, = supra=20 note 86, at 22.

94 Ken Sexton et al., Air = Pollution=20 Health Risks: Do Class and Race Matter?, 9 Toxicology & Indus. = Health=20 843, 851 (1993) (citing Inge F. Goldstein et al., Assessment of = Human=20 Exposure to Nitrogen Dioxide, Carbon Monoxide and Respirable = Particulates=20 in New York Inner-City Residences, 22 Atmospheric Env't 2127=20 (1988)).

95 Guidelines for = Carcinogen Risk=20 Assessment, 51 Fed. Reg. 33,992, 33,998 (1986).

96 EPA Equity Report , = supra note=20 12, at 36.

97 EPA Equity Report , = supra note=20 12, at 36.

98 Tim E. Aldrich & = Jack=20 Griffith, Environmental Epidemiology and Risk Assessment 215 = (Christopher=20 Cooke ed., 1993). Other estimates put the figure at about 63,000=20 chemicals. Thomas H. Maugh, II, Chemicals: How Many Are There?, = 199 Sci.=20 162 (1978).

99 15 U.S.C. = =A7=20 2603(e) (1988).

100 15 U.S.C. = =A7=20 2603(e)(1)(A) (1988). Of course, even such explicit = congressional=20 mandates beg the question regarding the initiation of a risk = assessment=20 because it is still not clear how the TSCA interagency testing = committee=20 will choose which substances or mixtures to evaluate.

101 Ellen Hall & = Barbara B.=20 Mandula, Formaldehyde, in Assessing Risks to Health , supra note = 20, at=20 57, 60 - 64.

102 Barbara L. Berney = & Barbara=20 B. Mandula, Tris, in Assessing Risks to Health, supra note 20, at = 91, 94 -=20 96.

103 Berney, supra note 24, = at 23=20 -25.

104 See infra part=20 III.B.3.

105 Dr. Bullard describes = this=20 problem as the "analysis-paralysis" syndrome, arguing that "as = long as the=20 problem is studied and restudied, no cleanup action will have to = be=20 taken." Bullard, supra note 53, at 261. Dr. Bullard has also = suggested=20 that the regulatory reliance on risk assessment may, in some = instances,=20 implicate Constitutional concerns. "Victims should not have to = prove that=20 they have been poisoned, that they are sick because of chemicals. = Waiting=20 for cause and effect is, in effect, the denial of equal = protection."=20 Lawlor, supra note 48.

106 The Southwest Network = for=20 Environmental and Economic Justice (SNEEJ) claims that,
EPA routinely factors politics and power into its = major=20 management decisions. An excellent example of this is the = comparison=20 between its decision to ban alar and its decision to ban = parathion. EPA=20 moved to cancel the use of the chemical alar on apples some = three weeks=20 after actress Meryl Streep testified before Congress....In = contrast, EPA=20 reached the conclusion that it should cancel the use of the = pesticide=20 parathion in 1987 because of the health threat to farmworkers, = but=20 withheld action until a staffmember leaked word of the coverup = last=20 year.
SNEEJ Comments, supra note 8, at = 93.

107 "The EPA, ... despite = their=20 environmental justice investigations, continues to rely heavily = upon=20 traditional risk assessment methodologies. For the federal = government to=20 effect meaningful change in the siting process and waste = management=20 generally, it must do more to incorporate citizens' risk = perceptions into=20 its risk calculations and its regulations." James S. Freeman & = Rachel=20 D. Godsil, The Question of Risk: Incorporating Community = Perceptions into=20 Environmental Risk Assessments, 21 Fordham Urb. L.J. 547 , 575 = (1994); see=20 also Clayton P. Gillette & James E. Krier, Risk, Courts and = Agencies,=20 138 U. Pa. L. Rev. 1027, 1032-33 (1990) (arguing that attitudes = about risk=20 must be understood before determining the appropriate = institutional=20 mechanism with which to regulate risk); Daniel Goleman, Hidden = Rules Often=20 Distort Ideas of Risk, N.Y. Times , Feb. 1, 1994, at C1, C10 = (suggesting=20 that discrepancies between public perceptions of risk and = scientific=20 quantification of risk may be attributable to the public's = tendency to=20 exaggerate involuntarily imposed, uncontrollable, inequitably = distributed,=20 man-made or catastrophic risks, even where those risks are not = objectively=20 greater than voluntarily accepted, equitably shared, controllable, = or=20 natural risks); infra note 224.

108 "[S]ome in the = environmental and=20 grassroots community argue that risk assessments are inherently = flawed=20 because of statistical and scientific failings in the risk = assessment=20 process, which results in an underestimation of the real risks." = Reath,=20 supra note 56 (quoting Linda Greer of the Natural Resources = Defense=20 Council); see also Deeohn Ferris, Remarks at the Children's = Environmental=20 Health Network Conference 1 (March 18 -19, 1994) (transcript on = file with=20 the N.Y.U. Environmental Law Journal) ("Another critical aspect of = protecting underserved communities is re-evaluating the science = industry=20 and data collection.").

109 There is growing = concern among=20 scientists, legal scholars, and environmental justice activists = that the=20 scientific methodology of risk assessment is itself causally = related to=20 disproportionately inadequate environmental protection. See, e.g., = Symposium on Health Research and Needs to Ensure Environmental = Justice:=20 Executive Summary & Proceedings and Recommendations (National=20 Institute of Environmental Health Sciences et al. eds., 1994) = (reporting=20 the proceedings and recommendations from the February 1994 = symposium which=20 involved seven Federal agency or department sponsors and over 1000 = participants, roughly half of which were government scientists and = officials, the remainder consisting of community representatives,=20 academics, and policy analysts); Robert D. Bullard & Beverly = H.=20 Wright, Environmental Justice for All: Community Perspectives on = Health=20 and Research Needs, 9 Toxicology & Indus. Health 821, 836 = (1993) ("In=20 order for any exposed population or contaminated community to get = help,=20 sweeping changes are needed in key areas of the science model and = the=20 environmental health research model."); Environmental Justice = Transition=20 Group, Recommendations to the Presidential Transition Team for the = U.S.=20 Environmental Protection Agency on Environmental Justice Issues 11 = (Dec.=20 21, 1992) (unpublished document, on file with the N.Y.U. = Environmental Law=20 Journal) [hereinafter Environmental Justice Recommendations ]; = Deeohn=20 Ferris, Testimony Before the Subcomm. on Civil and Constitutional = Rights=20 of the House Comm. on Judiciary 8 - 9 (Mar. 3, 1993) (unpublished=20 document, on file with the N.Y.U. Environmental Law Journal); Mary = H.=20 O'Brien, Poisoning the Poisoned: The Need to Use Our Multiple = Causation=20 Research and Eliminate Our Single Chemical Regulation, Address = Before the=20 National Institute of Environmental Health Sciences and the U.S.=20 Environmental Protection Agency (Feb. 10 -12, 1994) (unpublished = document,=20 on file with the N.Y.U. Environmental Law Journal).

110 This Student Article = is=20 concerned with outlining the theoretical possibility that = informational=20 biases within risk assessment disproportionately affect = communities of=20 color and poor communities, resulting in those communities = receiving less=20 than adequate environmental protection. There is, however, already = substantial empirical data showing that poor and minority = communities=20 experience relatively lower health status with respect to those = health=20 effects that are thought to be causally related to environmental=20 pollutants. See generally Laura E. Montgomery & Olivia = Carter-Pokras,=20 Health Status by Social Class and/or Minority Status: Implications = for=20 Environmental Equity Research, 9 Toxicology & Indus. Health = 729 (1993)=20 (presenting various studies that indicate that people who are = racial=20 minorities, or those with low income, have higher mortality rates = and=20 higher rates of most chronic conditions).

111 See generally Ken = Sexton et al.,=20 "Environmental Justice": The Central Role of Research in = Establishing a=20 Credible Scientific Foundation for Informed Decision Making, 9 = Toxicology=20 & Indus. Health 685, 713 (1993) ("For disparities in = environmental=20 health risks to occur by socioeconomic status or ethnicity/race, = these=20 demographic variables must be associated with systematic = differences in=20 (1) exposure to environmental agents, (2) susceptibility to the = effects of=20 environmental agents, or (3) both exposures and=20 susceptibilities.").

112 While some of these = distortions=20 have been acknowledged for quite some time, many are only now = receiving=20 scientific attention. For example, scientists have been aware of = the=20 problem of complex mixtures for many years. See generally Grose et = al.,=20 supra note 65 (recognizing the importance of complex mixtures in = achieving=20 a thorough understanding of human risk). Only recently, however,=20 scientists have observed that exposure to mixtures may be = demographically=20 correlated. See infra part II.A.

113 See infra part = II.A.1.

114 See infra part = II.A.2.

115 See infra part = II.A.3.

116 See infra part = II.A.4.

117 See generally EPA = Equity Report=20 , supra note 12, =A7 2.2.4 (presenting evidence that minority = communities=20 are systematically exposed to higher levels of certain air = pollutants).=20 Low income is positively correlated with high concentrations of = the six=20 "criteria" air pollutants regulated by the Clean Air Act: ozone, = nitrogen=20 dioxide, carbon monoxide, sulfur dioxide, particulate matter, and = lead.=20 Sexton et al., supra note 94, at 849 -50. "Criteria" air = pollutants are=20 established by EPA who then sets national minimum standards called = National Ambient Air Quality Standards (NAAQS). 42 U.S.C. = =A7=20 7409 (1988). Similarly, minorities are more likely to live in = areas=20 not in compliance with EPA requirements for criteria pollutants. = Sexton et=20 al., supra note 94, at 849 -50. A 1991 study found that one-third = of all=20 Whites lived in nonattainment areas for carbon monoxide, while 46% = of all=20 African Americans and 57% of all Hispanics lived in carbon = monoxide=20 nonattainment areas. Sexton et al., supra note 94, at 850. Initial = data=20 also suggest that minority and low- income people experience a = higher=20 exposure to indoor air pollutants. Sexton et al., supra note 94, = at=20 851.

118 While the demographic = data for=20 exposure to water pollutants is less developed than that of air=20 pollutants, there are sufficient indicators to warrant concern = that poor=20 and minority communities are disproportionately exposed to water=20 pollutants. See Rebecca L. Calderon et al., Health Risks from = Contaminated=20 Water: Do Class and Race Matter?, 9 Toxicology & Indus. Health = 879,=20 894 - 95 (1993) ("Despite the sparseness and limitations of the = data, the=20 existing data suggest that environmental inequities exist.") = People such=20 as migrant farm workers who experience primitive living = conditions, for=20 example, are more likely to be exposed to water contaminants. A = 1991 study=20 found extremely high exposure levels to various contaminants in = the water=20 supply of migrant farms in North Carolina. Stephen Cieselski et = al., The=20 Microbiologic Quality of Drinking Water in North Carolina Migrant = Labor=20 Camps, 81 Am. J. Pub. Health 762, 762- 63 (44% total coliforms, = 26% fecal=20 coliforms, and 20% E. Coli). The same study found none of these = organisms=20 in water supplies of surrounding businesses and residences. Id. at = 762.=20 Migrant farm workers are mostly Hispanic (75%) and Native American = (20%).=20 Donna M. Arbab & B. Louise Weidner, Infectious Diseases and = Field=20 Water Supply and Sanitation Among Migrant Farm Workers, 76 Am. J. = Pub.=20 Health 694, 694 (1986).

119 Exposure to hazardous = wastes is=20 highly correlated to racial and economic criteria. A number of = studies=20 dating back to 1983 demonstrate that both race and income are = reliable=20 predictors of proximity to hazardous waste facilities. See, e.g., = Mohai=20 & Bryant, supra note 2 (reviewing 18 studies and concluding = that race=20 is a factor independent of class in the disproportionate siting of = hazardous facilities). Individual case studies are equally = revealing. For=20 example, in South Central Los Angeles, in a zip code area known as = the=20 "dirtiest" in California, where 33 million pounds of waste = chemicals were=20 discharged in 1989, the population is 59% African American and 38% = Latino.=20 Bullard & Wright, supra note 109, at 825; see also Magdi R. I. = Soliman=20 et al., Hazardous Wastes, Hazardous Materials and Environmental = Health=20 Inequity, 9 Toxicology & Indus. Health 901, 904 (1993) = ("[R]esearchers=20 have documented that hazardous waste sites are disproportionately = located=20 near African-American Communities and other minorities including = Hispanics=20 and Native Americans.").

120 See generally Marion = Moses et=20 al., Environmental Equity and Pesticide Exposure, 9 Toxicology = &=20 Indus. Health 913, 914 ("People of color and low- income groups = bear a=20 disproportionate share of the potential health risks from exposure = to=20 pesticides.").

121 Mohai & Bryant, = supra note=20 2, at 166 - 67 ("Where the distribution of pollution has been = analyzed by=20 both income and race (and where it was possible to weigh the = relative=20 importance of each), in most cases race has been found to be more = strongly=20 related to the incidence of pollution.").

122 William H. Hallenbeck = &=20 Kathleen M. Cuningham, Quantitative Risk Assessment For = Environmental and=20 Occupational Health 3 (1986).

123 Id.

124 10 -6'] x 100 =3D 10 = -4']. "Thus,=20 an acceptable lifetime risk of 10 -6'] may not actually be as = conservative=20 as intended." Id. This calculation is a simplification which = assumes no=20 overlap in the health effects caused by those 100 substances. It = is=20 important to note that the additivity critique is largely a risk=20 management and risk communication problem. The risk management = question is=20 whether decisionmakers should increase the stringency of = regulations with=20 respect to groups that face more individual substances. The risk=20 communication question concerns the propriety of presenting single = substance risk assessments to a community that may be exposed to = multiple=20 substances.

125 Bullard, supra note = 53, at 254=20 -55.

126 The term "toxic = doughnut" is a=20 phrase coined by community organizer Hazel Johnson. While = originally used=20 to refer to Altgeld Gardens, the phrase has been increasingly = applied by=20 environmental justice activists to describe similar situations = throughout=20 the country. See, e.g., Deeohn Ferris, Testimony before the = Subcomm. on=20 Superfund, Recycling, and Solid Waste Management of the Senate = Comm. on=20 Environment and Public Works (July 28, 1993) (unpublished = document, on=20 file with the N.Y.U. Environmental Law Journal).

127 Greenpeace, Home = Street, USA,=20 Greenpeace Mag. , Oct./Nov./Dec. 1991, at 8, 8 -13.

128 Bullard & Wright, = supra note=20 109, at 835.

129 Calabrese has observed = that,
[w]hile nearly the entire thrust of public health risk = assessment activities has involved derivations for individual = compounds,=20 all agree that the real world involves multiple chemical = exposures,=20 either concurrently or sequentially. Despite universal agreement = on=20 this, regulatory agencies, especially in the environmental = domains, have=20 been slow to directly address and specifically incorporate the = knowledge=20 of interactions into the risk assessment process. =
Calabrese,=20 Multiple Chemical Interactions , supra note 63, at 601.

130 Daniel Krewski et al., = Carcinogenic Risk Assessment of Complex Mixtures, in Health = Hazards and=20 Risks from Exposure to Complex Mixtures and Air Toxic Chemicals = supra note=20 65, at 147, 151.

131 Guidelines for the = Health Risk=20 Assessment of Chemical Mixtures, 51 Fed. Reg. 34,014, 34,015=20 (1986).

132 Calabrese, Multiple = Chemical=20 Interactions , supra note 63, at 13 -14.

133 Calabrese, Multiple = Chemical=20 Interactions , supra note 63, at 14.

134 Calabrese, Multiple = Chemical=20 Interactions , supra note 63, at 14.

135 For example, = consumption of=20 alcohol is thought to enhance the toxicity of carbon disulfide, an = industrial solvent. Calabrese, Multiple Chemical Interactions , = supra note=20 63, at 487- 88.

136 Synergism occurs when, = for=20 example, "two compounds, innocuous by themselves, might interact = at low=20 doses to form a new substance that is toxic." National Research = Council ,=20 Pesticides in the Diets of Infants and Children 12 = (1993).

137 Guidelines for the = Health Risk=20 Assessment of Chemical Mixtures, 51 Fed. Reg. 34,014, 34,015=20 (1986).

138 Id.

139 Id.

140 See, e.g., Robert R. = Vanderslice=20 et al., Problems in Assessing the Risks of Mixtures of = Contaminants in=20 Drinking Water, in Health Hazards and Risks from Exposure to = Complex=20 Mixtures and Air Toxic Chemicals , supra note 65, at 117 = (discussing=20 potential synergistic effects among drinking water=20 contaminants).

141 51 Fed. Reg. 34,014, = 34,016=20 (1986); see also Krewski et al., supra note 130, at 157 ("The = limited=20 number of laboratory studies on joint exposure to chemical = carcinogens,=20 however, has not provided a clear indication of the extent to = which=20 synergistic effects among carcinogens may occur.").

142 "Perhaps the most = important=20 complication [of evaluating risks of exposure to chemical = mixtures] is the=20 potential for interaction among the mixture's constituents, = including=20 synergistic effects in which the combined effect of two or more = substances=20 is greater than the sum of the effects of each agent alone." = Krewski et=20 al., supra note 130, at 147.

143 42 U.S.C. = =A7=20 9604(i)(5)(A) (1988).

144 51 Fed. Reg. 34,014, = 34,015=20 (1986).

145 Id. at 34,016 (Table=20 1).

146 Id. at = 34,021.

147 Id. EPA cites a = classic 1979=20 study by E.C. Hammond et al. as an example of a synergistic = effect. E.C.=20 Hammond et al., Asbestos Exposure, Cigarette Smoking and Death = Rates, 330=20 Annals N.Y. Acad. Sci. 473 (1979) (finding that the relative risk = of lung=20 cancer attributable to smoking was 11, to asbestos exposure was 5, = and to=20 the combination was 53).

148 Rita S. Schoeny & = Elizabeth=20 Margosches, Evaluating Comparative Potencies: Developing = Approaches to=20 Risk Assessment of Chemical Mixtures, in Health Hazards and Risks = from=20 Exposure to Complex Mixtures and Air Toxic Chemicals , supra note = 65, at=20 125, 125.

149 "The Agency and its = reviewers=20 agree that as the number of compounds in the mixture increases, an = assumption of additivity will become less reliable in estimating = risk." 51=20 Fed. Reg. 34,014, 34,024 (1986).

150 See supra notes=20 117-21.

151 See Mohai & = Bryant, supra=20 note 2.

152 See supra notes 125=20 -28.

153 See, e.g., Patrick C. = West et=20 al., Minority Anglers and Toxic Fish Consumption: Evidence from a=20 Statewide Survey of Michigan, in Race and the Incidence of = Environmental=20 Hazards: A Time for Discourse , supra note 1, at 100; see also EPA = Equity=20 Report , supra note 12, at 12-15.

154 EPA Equity Report , = supra note=20 12, at 12.

155 Edward J. Calabrese , = Pollutants=20 and High-Risk Groups: The Biological Basis of Increased Human=20 Susceptibility to Environmental and Occupational Pollutants 150 = (1978)=20 [hereinafter Calabrese, High-Risk Groups ] ("[Pollutants] are also = known=20 to concentrate in the fat tissue of various freshwater fish = species such=20 as salmon and trout ... People following certain ethnic diets ... = which=20 involve consuming large amounts of fish as compared with the = general=20 population would probably be exposed to higher levels of=20 [pollutants].").

156 West et al., supra = note 153. The=20 Michigan study surveyed 2600 registered anglers by mail for total = fish=20 consumption and received a 47.3% response. West et al., supra note = 153, at=20 101.

157 West et al., supra = note 153, at=20 102.

158 West et al., supra = note 153, at=20 110.

159 West et al., supra = note 153, at=20 110 -12.

160 See supra notes 68 -71 = and=20 accompanying text.

161 See, e.g., Paul A. = Locke,=20 Reorienting Risk Assessment, Envtl. F. , Sept.-Oct. 1994, at 29 = (arguing=20 that current risk assessment practices do not adequately account = for=20 non-carcinogenic health effects and that environmental health = regulation=20 suffers as a result).

162 See Ferris, supra note = 108, at=20 2.

163 Latin, supra note 59, = at 98 -100=20 (stating that estimates produced by different extrapolative models = differ=20 by orders of magnitude and lack an experimental basis for choosing = between=20 them). As discussed above, supra note 81, while extrapolation may=20 overstate risk, this possibility does not undermine an = environmental=20 justice critique of risk assessment.

164 S. Rep. No. 1196, 91st = Cong., 2d=20 Sess. 10 (1970); Lead Indus. Ass'n v. EPA, 647 F.2d 1130, 1152-53, = 1156 -=20 60 (D.C. Cir.), cert. denied, 449 U.S. 1042 (1980); see also = Robert Frank,=20 Susceptibility to Air Pollution: The Clean Air Act, in Variations = in=20 Susceptibility to Inhaled Pollutants 443 (Joseph D. Brain et al. = eds.,=20 1988) (discussing the legal authority and efficacy of accounting = for=20 subgroup susceptibility in risk assessment); John G. Harkins, Jr. = et al.,=20 A Legal Viewpoint on the Relevance of Phenotypic Variation to Risk = Assessments, in Phenotypic Variation in Populations: Relevance to = Risk=20 Assessment 213 (Avril D. Woodhead et al. eds., 1986) = (same).

165 EPA Equity Report , = supra note=20 12, at 33 -34.

166 EPA Equity Report, = supra note=20 12, at 33 -34.

167 For instance, in its=20 Environmental Equity report, EPA stated that
[w]hen the Agency does have information about the=20 susceptibility of certain subgroups in the population, this = information=20 is taken into account in the risk assessment. For example, = children have=20 been specially singled out as being at risk from exposure to = lead. The=20 elderly are sensitive to carbon monoxide, and particulate matter = exposure. Asthmatics are a sensitive subgroup for sulfur oxides. = There=20 are other examples where health information for special = population=20 groups is key to the risk assessment finding and examples where = such=20 focus is not achievable.
EPA Equity Report , supra = note 12,=20 at 33 -34.

168 EPA Equity Report, = supra note=20 12, at 33 -34.

169 See generally Edward = J.=20 Calabrese, Ecogenetics: Genetic Variation in Susceptibility to=20 Environmental Agents (1984) [hereinafter Calabrese, Ecogenetics ]=20 (analyzing the effect of genetic factors on the outcome of = environmentally=20 induced disease); Calabrese, High-Risk Groups , supra note 155=20 (identifying and quantifying hypersusceptible population segments = with=20 respect to pollutant toxicity and biological factors, genetic = disorders,=20 nutritional deficiencies, disease processes, and behavior); = Richard Rios=20 et al., Susceptibility to Environmental Pollutants Among = Minorities, 9=20 Toxicology & Indus. Health 797 (1993) (examining biological=20 susceptibility of minorities to environmental pollutants in = general and=20 also providing specific examples).

170 Of course, for many of = these=20 variables the appropriate response is not to fix the risk = assessment, but=20 rather to fix the underlying susceptibility. See, e.g., Calabrese, = High-Risk Groups, supra note 155, at 174 -75 (suggesting = nutritional=20 supplementation programs to increase adaptive capacity to = environmental=20 pollutants).

171 Calabrese , = Ecogenetics , supra=20 note 169, at 29.

172 Calabrese , = Ecogenetics , supra=20 note 169, at 29.

173 Calabrese , = Ecogenetics , supra=20 note 169, at 18 -27, 323 ("G- 6 -PD deficiency has a demonstrable = causal=20 history of enhancing one's susceptibility to industrial=20 pollutants.").

174 Calabrese, High-Risk = Groups ,=20 supra note 155, at 197 ("A clear illustration of this situation = [high-risk=20 groups] is seen in cities that have a high ozone concentration and = a large=20 number of blacks with the G- 6 -PD deficiency.").

175 Calabrese, High-Risk = Groups ,=20 supra note 155, at 197.

176 Calabrese, Ecogenetics = , supra=20 note 169, at 30.

177 Calabrese, Ecogenetics = , supra=20 note 169, at 41 (emphasizing that despite suspicion, the evidence = with=20 respect to the increased susceptibility of those with sickle-cell = trait is=20 inconclusive).

178 Calabrese, Ecogenetics = , supra=20 note 169, at 326.

179 Calabrese, Ecogenetics = , supra=20 note 169, at 290.

180 Rios et al., supra = note 169, at=20 802. The very existence of chemically induced susceptibility, not = to=20 mention its prevalence, is widely controverted. See e.g., American = Council=20 on Science and Health, MCS: Multiple Chemical Sensitivity 27 = (concluding=20 that multiple chemical sensitivity is scientifically unsupported,=20 unrecognized by the mainstream medical community and largely based = on=20 "junk" science). But see William Rea, Chemical Sensitivity: = Principles and=20 Mechanisms (1992) (providing a detailed model supporting the = theory of=20 multiple chemical sensitivity); Linda Lee Davidoff, Multiple = Chemical=20 Sensitivities, Amicus J. (Nat. Resources Def. Council), Fall 1989, = at 12=20 (discussing the possibility that multiple chemical sensitivity is = widely=20 prevalent). For an excellent, more balanced, discussion of the = possibility=20 that chemical exposure may suppress the body's immune system and = thus=20 increase susceptibility, see Michael I. Luster et al., Chemical = Pollutants=20 and "Multiple Chemical Sensitivities," in Phantom Risk: Scientific = Inference and the Law 379 (Kenneth R. Foster et al. eds., 1993)=20 (presenting data showing the existence of chemically induced = sensitivities=20 while also articulating the tremendous scientific uncertainties=20 surrounding environmentally related hypersensitivities in=20 general).

181 Rios et al., supra = note 169, at=20 802 (citing George Friedman-Jimenez, Occupational Disease Among = Minority=20 Workers: A Common and Preventable Public Health Problem, 37 Am. = Ass'n=20 Occupational Health Nurses J. 64 (1989)).

182 See generally A.P. = Polednak,=20 Racial and Ethnic Differences in Disease (1989) (discussing = variations in=20 disease frequencies occurring among various racial, ethnic and=20 socioeconomic groups); Robert W. Miller, Epidemiologic Evidence = for=20 Genetic Variability in the Frequency of Cancer: Ethnic = Differences, in=20 Phenotypic Variations in Populations: Relevance to Risk Assessment = , supra=20 note 164, at 65 (same).

183 Rios et al., supra = note 169, 808=20 - 10.

184 Rios et al., supra = note 169, at=20 808 (citing U.S. Dep't of Health & Human Services, Report of = the=20 Secretary's Task Force on Black and Minority Health 1, 4=20 (1986)).

185 Rios et al., supra = note 169, at=20 808 (discussing some studies which indicate that exposure to = cadmium may=20 induce high blood pressure, exacerbating health problems caused by = hypertension) (citing Henry A. Schroeder, Cadmium as a Risk Factor = in=20 Hypertension, 18 J. Chron. Dis. 647 (1965) and Henry A. Schroeder, = The=20 Role of Trace Metals in Cardiovascular Diseases, 58 Med. Clinics = N. Am.=20 381 (1974)).

186 Rios et al., supra = note 169, at=20 808.

187 Rios et al., supra = note 169, at=20 809; see also Calabrese, Ecogenetics , supra note 169, at 243 - 44 = (describing study on rats which demonstrated that a diabetic state = increases the hepatotoxicity of organic solvents).

188 Rios et al., supra = note 169, at=20 809 (citing Indian Health Services, U.S. Dep't of Health & = Human=20 Services, Trends in Indian Health 1991 (1991); U.S. Dep't of = Health &=20 Human Services, Health Status of Minorities and Low Income Groups = (3d ed.=20 1991); and U.S. Dep't of Health & Human Services, Report of = the=20 Secretary's Task Force on Black and Minority Health 1, 7=20 (1986)).

189 Rios et al., supra = note 169, at=20 809 (citing Calabrese, Ecogenetics , supra note 169).

190 Rios et al., supra = note 169, at=20 809 (citing Centers for Disease Control, Tuberculosis in Blacks, = 36=20 Morbidity & Mortality Wkly. Rep. 212 (1987); Centers for = Disease=20 Control, Tuberculosis -- United States, 1985, 35 Morbidity & = Mortality=20 Wkly. Rep. 699 (1986); and J.A. Jereb et al., Tuberculosis = Morbidity in=20 the United States: Final Data, 1990, 40 Morbidity & Mortality = Wkly.=20 Rep. 23 (1991)).

191 Rios et al., supra = note 169, at=20 809 (citing Centers for Disease Control, Asthma -- United States, = 1980=20 -1987, 39 Morbidity & Mortality Wkly. Rep. 493 (1990)) = .

192 Joan Borysenko,=20 Psychophysiological Variables, in Variations in Susceptibility to = Inhaled=20 Pollutants , supra note 164, at 295, 298 -301 (arguing that = "psychological=20 factors affected by the psychosocial climate must be considered as = seriously as biological risk factors in predicting disease=20 susceptibility").

193 Rios et al., supra = note 169, at=20 813 (citing U.S. Dep't of Health & Human Services, Health = Status of=20 Minorities and Low Income Groups (3d ed. 1992)).

194 Rios et al., supra = note 169, at=20 813 (citing Centers for Disease Control, Tuberculosis in Blacks, = 36=20 Morbidity & Mortality Wkly. Rep. 212 (1987)).

195 See, e.g., Jean Mayer, = Hunger=20 and Undernutrition in the United States, 120 J. Nutrition 919, 920 = -22=20 (1990) .

196 Calabrese, High-Risk = Groups ,=20 supra note 155, at 93 -114; see also Laurence N. Kolonel, = Variability in=20 Diet and its Relation to Risk in Ethnic and Migrant Groups, in = Phenotypic=20 Variation in Populations: Relevance to Risk Assessment , supra = note 164,=20 at 129 (suggesting that diet accounts for a large part of the = observed=20 variations in cancer incidence among ethnic groups in = Hawaii).

197 Calabrese, High-Risk = Groups ,=20 supra note 155, at 96 -101 (pointing out that the subgroup = prevalence of=20 vitamin C deficiency is uncertain, though at least one study = suggests that=20 low income is positively correlated with such a = deficiency).

198 Calabrese, High-Risk = Groups ,=20 supra note 155, at 191- 92.

199 Rios et al., supra = note 169, at=20 814.

200 Rios et al., supra = note 169, at=20 811 (citing U.S. Dep't of Health & Human Services, Health = Status of=20 Minorities and Low Income Groups (3d ed. 1991)).

201 Rios et al., supra = note 169, at=20 811 (citing Edward J. Calabrese, Age and Susceptibility to Toxic=20 Substances (1986); and Calabrese, High-Risk Groups, supra note = 152); A.=20 Jane Warren & Shelley Weinstock, Age and Preexisting Disease, = in=20 Variations in Susceptibility to Inhaled Pollutants , supra note = 192, at=20 253.

202 See, e.g., Ronald M. = Davis,=20 Current Trends in Cigarette Advertising and Marketing, 316 New = Eng. J.=20 Med. 725, 728 -30 (1987) (finding that the marketing efforts of = several=20 cigarette companies have increasingly targeted African American = and=20 Hispanic communities).

203 Rios et al., supra = note 169, at=20 810 -11 (citing U.S. Dep't of Health & Human Services, Health = Status=20 of the Disadvantaged: Chartbook 1990 (1990)).

204 Rios et al., supra = note 169, at=20 810 -11; Calabrese, High-Risk Groups , supra note 155, at 135=20 -50.

205 See supra notes 105 = -08 and=20 accompanying text.

206 See, e.g., Sierra = Club, The=20 Sierra Club 1994 Congressional Platform (1994) (urging legislators = to=20 resist measures that would require risk assessments because such = measures=20 would result in unjustified delay and expense); Mary O'Brien, = Facing Down=20 Pesticide Risk Assessment, Global Pesticide Campaigner (Pesticide = Action=20 Network, San Francisco, Cal.) Mar. 1994, at 1, 1 (1994) = (advocating for=20 the rejection of risk assessment in the pesticide context because = it=20 establishes acceptable levels of exposure without seeking the = consent of=20 those exposed and without assessing nonpesticide = alternatives).

207 See supra part = I.A.

208 See supra note = 9.

209 This point, of course, = is at the=20 heart of the risk assessment controversy. See supra part I.A.3. = Some=20 observers take this point further and argue that the complete = abandonment=20 of risk assessment would aggravate rather than ameliorate = environmental=20 justice concerns. See Albert L. Nichols, Risk- Based Priorities = and=20 Environmental Justice, in Worst Things First? The Debate Over = Risk-Based=20 National Environmental Priorities , supra note 53, at 267, 267- 68 = (arguing that without risk assessment as a key element to = regulatory=20 decisionmaking, "political power would be even more important than = it is=20 today in determining which problems get addressed and where = polluting=20 facilities are located").

210 See supra parts I.B.2, = I.B.3.=20 But see O'Brien, supra note 206, at 17 ("Do we really need more = and better=20 risk assessments for deciding how much of each pesticide active = ingredient=20 shall be considered legally 'safe' in children's food? ... = Ultimately,=20 children are not protected by better risk assessments of the = pesticides=20 ... they are protected by altering the practices of those who use = the=20 pesticides.").

211 Of course, this = argument does=20 not counter the other environmental justice objections to risk = assessment=20 such as the possible delay in regulatory action, the failure to=20 incorporate community perceptions of risk, and the fear that risk=20 management decisionmaking is biased. See supra notes 105 -108 and=20 accompanying text.

212 See supra part = II.

213 Environmental Justice=20 Recommendations , supra note 109, at 11. See also Ferris, supra = note 108,=20 at 2 ("Any meaningful analysis intended to protect underserved = communities=20 will recognize that multiple-cumulative-combination exposures are=20 occurring in underprotected communities and effects associated = with these=20 types of exposure must be a central focus of the = research.").

214 EPA Equity Report , = supra note=20 12, at 31. EPA was inconsistent in its use of income as a variable = that=20 may increase exposure and susceptibility. The above recommendation = unfortunately fails to include income, while the immediately = following=20 recommendation discusses the need to correlate exposure and risk = levels=20 with U.S. Census data in order to "permit quantitative analysis of = the=20 proportionality of exposures and risk according to demographic=20 classifications of race, ethnicity, gender, age and income." EPA = Equity=20 Report , supra note 12, at 31.

215 SNEEJ Comments, supra = note 8, at=20 94 - 96, 102 (arguing that EPA's use of language reveals its lack = of=20 commitment toward remedying data gaps within risk assessments). = One could=20 further object to EPA's use of such phrases as "in certain cases" = and "it=20 may be helpful" as qualifications of the need to gather = demographic=20 information.

216 A recent memo from = Robert M.=20 Sussman, Chair of EPA's Science Policy Council (SPC), to Carol = Browner,=20 Administrator, demonstrates the Agency's awareness of the = informational=20 bias within risk assessment. EPA Science Policy Council, Report of = the EPA=20 Science Policy Council on Addressing "Science and Judgment in Risk = Assessment," A Report by the National Research Council, in Inside = EPA's=20 Clean Air Report , June 30, 1994, at 19, 19 -23. Two of the SPC's = eight=20 thematic areas for improving regulatory risk assessment were the=20 incorporation of multi-path and multi-source exposures and = incorporation=20 of "inter-individual" susceptibilities. Furthermore, the memo = highlights=20 the fact that these improvements are particularly important to = alleviate=20 environmental justice concerns. See also EPA Waste Program = Developing Tool=20 to Expand Risk Consideration, Inside EPA's Risk Policy Report , = Oct. 14,=20 1994, at 6, 6 -7 (detailing an effort by EPA's Office of Solid = Waste to=20 perform sophisticated multi-pathway analyses of exposure to = particular=20 substances).

217 Sexton et al., supra = note 111,=20 at 722; see also Yolanda Banks Anderson et al., Overview of the=20 EPA/NIEHS/ATSDR Workshop - "Equity in Environmental Health: = Research=20 Issues and Needs," 9 Toxicology & Indus. Health 679 (1993)=20 (summarizing eleven specific research recommendations necessary to = better=20 understand the distribution of environmental health risks across = groups);=20 supra note 109 (documenting the growing concern with improving = government=20 science).

218 See supra notes 164=20 -68.

219 One of Justice = Breyer's examples=20 of unreasonable protection is the cleanup standards for a swamp in = an=20 isolated area that would require it to be so clean that a = hypothetical=20 child could eat the soil there for 70 years without suffering = negative=20 health effects. Breyer, supra note 20, at 11-12.

220 Sexton et al., supra = note 94, at=20 847.

221 See, e.g., 42 U.S.C. = =A7=A7=20 7409(b)(2), 7602(h) = (1988=20 & Supp. IV 1992) (stating the requirements under the Clean Air = Act=20 that national secondary ambient air quality standards protect = public=20 welfare from known or anticipated adverse effects, including = effects on,=20 "soils, water, crops, vegetation, manmade materials, animals, = wildlife,=20 weather, visibility, and climate").

222 Also, where there is = scientific=20 uncertainty with respect to adverse effects to public health or = the=20 environment, risk assessment would continue to employ conservative = assumptions in order to set standards which adequately protect = against=20 risk.

223 See, e.g., Breyer, = supra note=20 20, ch. 1. Of course, such an easy formulation will not satisfy = many=20 critics. Aaron Wildavsky, for example, who posits that the cost of = regulations will itself reduce public safety, argues that = government=20 protection is only appropriate when risk is high and certainty is = high.=20 Aaron Wildavsky, Richer is Safer, 60 The Public Interest 23, 25 = (1980).=20 Wildavsky's thesis requires us to ask upon which group should we = impose=20 the incentive to increase certainty. If we assume no government = action=20 will occur until both risk and certainty are high, then those most = exposed=20 to regulatable substances will shoulder the incentive to increase=20 certainty. Conversely, if we require government action until risk = is low=20 and certainty is high, then those regulated will shoulder the = incentive to=20 increase certainty. There are strong policy arguments for = preferring the=20 latter presumption including the correlation between race and = income and=20 exposure, and the relative political and economic positions of = those who=20 are potentially exposed to substances versus those who are = potentially=20 regulated.

224 This Student Article = deals only=20 with a proposal to involve the public in the risk assessment = process, as=20 opposed to the very different proposal to involve public = perceptions of=20 risk in the risk management process. The latter proposal would = require=20 regulatory bodies to consider in their decisionmaking process the = affected=20 community's sense of, inter alia, control, priorities, fear, and=20 suspicion. See generally Ellison Folk, Public Participation in the = Superfund Process, 18 Ecology L.Q. 173, 189 (1991) ("Whether a = particular=20 community feels that a risk is acceptable is a political and = social=20 decision as much as a technical and rational one."); Freeman & = Godsil,=20 supra note 107 (discussing the potentially positive role of public = participation in the risk management process as it relates to the = siting=20 of hazardous facilities); Michael Gerrard, Fear and Loathing in = the Siting=20 of Hazardous and Radioactive Waste Facilities: A Comprehensive = Approach to=20 a Misperceived Crisis, 68 Tul. L. Rev. 1047, 1148 - 49 (1994) = (stating=20 that a community's ability to participate in the siting and = management of=20 a facility is an important factor in increasing public acceptance = of the=20 facility); Gillette & Krier, supra note 107 (discussing the = advantages=20 and disadvantages of an increased public role in risk management); = Thomas=20 O. McGarity, Public Participation in Risk Regulation, 1 Risk 103 = (1990)=20 (providing six models for public participation in risk=20 regulation).

225 See Environmental = Health Network=20 and the National Toxics Campaign Fund , Inconclusive By Design: = Waste,=20 Fraud and Abuse in Federal Environmental Health Research vii = (1992) ("One=20 of the most heinous and pervasive defects of the health assessment = program=20 has been a lack of ATSDR contact with local residents. The result = has been=20 an appalling absence of even the most basic understanding of local = conditions."); U.S. General Accounting Office , Superfund: Public = Health=20 Assessments Incomplete and of Questionable Value (1991).

226 EPA Equity Report, = supra note=20 12, at 19.

227 Public participation = is also=20 critical in order to build trust within affected communities. This = trust=20 in turn may positively impact the effectiveness of EPA programs. = EPA,=20 Superfund Administrative Improvements, Final Report 31-33 (1993) = ("A=20 critical problem in the Superfund program is the lack of support = for the=20 cleanup among the communities around Superfund sites ... = [therefore]=20 [p]ublic participation is critical to the effective and speedy=20 implementation of cleanups.").

228 See supra part = I.C.

229 See supra notes = 101-103 and=20 accompanying text.

230 42 U.S.C. = =A7=20 9617(e)(1) (1988 & Supp. IV 1992).

231 At present, TAGs are = issued for=20 up to $50,000. The President may waive the $50,000 limitation = where such a=20 waiver is necessary to carry out the purposes of the subsection. = 42 U.S.C. = =A7=20 9617(e)(2) (1988 & Supp. IV 1992).

232 See generally Richard = L. Hembra,=20 U.S. General Accounting Office, EPA's Superfund TAG Program: = Grants=20 Benefit Citizens but Administrative Barriers Remain (1992) = (discussing=20 problems of access to TAG grants including the complicated = application=20 procedures).

233 Barbara L. Berney = & Jack=20 Needleman, Evaluating Carcinogenic Risks, in Assessing Risks to = Health ,=20 supra note 20, at 251, 251-54; see also Needleman, supra note 38, = at 14=20 -17 (distinguishing those risk assessments performed in regulatory = agencies and those performed in educational agencies).

234 Berney & = Needleman, supra=20 note 233, at 252.

235 Berney & = Needleman, supra=20 note 233, at 252.

236 Berney & = Needleman, supra=20 note 233, at 252; see also Barbara L. Berney & Ellen Hall, = Passive=20 Smoking and Lung Cancer, in Assessing Risks to Health , supra note = 20, at=20 185, 191-206 (describing OSH's health risk assessment of passive = smoking=20 performed in the 1980s); Jack Needleman, Relating Risk Assessment = to Risk=20 Management, in Assessing Risks to Health , supra note 20, at 301, = 303=20 (concluding that educationally-oriented agencies often prefer = qualitative=20 to quantitative risk assessments).

237 Berney & = Needleman, supra=20 note 233, at 252. This tendency is probably caused by three = contributing=20 factors: the fact that clear data underlying a regulatory risk = assessment=20 is required in order to defend against contesting parties; the = increased=20 cost to society associated with regulation as opposed to = education; and=20 the immediate need to prioritize regulatory efforts in order to = comply=20 with congressional mandates.

238 Berney & = Needleman, supra=20 note 233, at 252.

239 Furthermore, "[t]his = flexibility=20 to act on findings that are perhaps a little short of conclusive = is=20 important to the public health; to wait for the last bit of = evidence may=20 be to allow unnecessary illness and deaths in the interim." Berney = &=20 Needleman, supra note 233, at 252.

240 Alternatively, this = analysis may=20 speak toward the need for a centralized risk assessment body that = could=20 have as its mandate the initiation of a certain number of risk = assessments=20 in response to inconclusive public perceptions. For other benefits = of a=20 centralized risk assessment agency, see Breyer , supra note 20, = ch.=20 3.

241 Sexton et al., supra = note 94, at=20 847 ("Generally, the cost of estimating exposure is directly = proportional=20 to the certainty of the estimate.").

242 Dr. Bullard is an = active=20 proponent of shifting the burden of proving safety to producers of = risks.=20 "The environmental justice strategy would require the entities = that are=20 applying for operating permits for landfills, incinerators, = smelters,=20 refineries, chemical plants, and so forth to prove that their = operations=20 are not harmful to human health, will not disproportionately = impact racial=20 and ethnic minorities and other protected groups, and are=20 nondiscriminatory." Bullard, supra note 53, at 248.

243 42 U.S.C. = =A7=20 7503 (1988 & Supp. IV 1992).

244=20
A poignant example of the difference between an EPA = risk=20 assessment and the public's perception of risk comes from the = Superfund=20 site in Oroville, California. A resident of the community, who = lived two=20 miles from one of the sites when it was listed on the NPL, told = me: 'I'm=20 sick, I live in a cancer cluster, my well's contaminated with=20 pentachlorophenol, and the EPA tells me the water is safe to = drink.'=20
Folk, supra note 224, at 189; see also Lavelle & = Coyle,=20 supra note 4 (profiling numerous communities where EPA enforcement = at=20 Superfund sites appears to be inadequate to protect human=20 health).

245 James A. Rogers & = David H.=20 Topol, Proposals for Reforming Superfund, C883 A.L.I.-A.B.A. 1, 15 = (1994).

246 See Breyer , supra = note 20, at=20 17-18; see also Superfund Health Agency Expands Human Exposure = Monitoring,=20 Inside EPA's Risk Policy Report , Nov. 18, 1994, at 19, 19 -20 = (describing=20 the Agency for Toxic Substances & Disease Registry's = initiative to=20 improve exposure assessments at Superfund sites through the use of = direct=20 biological and environmental monitoring and predicting that there = may be=20 many cases where actual exposure is less than currently assumed).=20 =




------=_NextPart_000_0000_01C619F5.950D5A40--