GiveBetter x Los Angeles

Communities for a Better Environment

Statewide environmental justice organizing, science and legal advocacy

Research time: 10 min on GPT-6 Astra Light
  • Research — reviewed programs, finances and impact evidence.

Updated: 2026-09-14

Donate

Summary

What do they do? CBE combines community organizing, scientific research and litigation to reduce pollution in California communities. Its work includes Exide cleanup and refinery disputes in Los Angeles, alongside substantial Bay Area and statewide activity. A judgment-based partial-health central estimate is about $75.5 million per 10 LA/Orange QALYs; it is not an observed return or a donor-ready recommendation.

Why we’re interested in this organization:

  • Current primary documents identify concrete cleanup barriers, including access to multi-family properties.

  • Legal, scientific and resident knowledge can complement government cleanup capacity.

  • Original financial returns identify the full recipient cost rather than only a selected campaign budget.

Our main reservations:

  • Policy probability, exposed population, dose reduction and marginal funding response are judgment inputs, with no verified next-dollar work plan; zero or harm remains possible.

  • The effect of remaining soil remediation on local blood lead and clinical outcomes is not measured here.

  • Statewide activity and coalition overlap prevent allocating all recipient benefits to Los Angeles.

What do you get for your dollar? $75.5M per better life: ten additional quality-adjusted life years in Los Angeles. Judgment-based partial-health; full recipient expenses; not all-resource social cost.

Additional annual ambient PM2.5 exposure reduction among an explicitly modeled older cohort; finite discounted all-cause survival differences, without adding overlapping mortality/morbidity or lead/IQ units.

1. What do they do?

CBE, EIN 94-2998086, operates in Southeast Los Angeles and Wilmington as well as East Oakland and Richmond. Ordinary support funds organizing, science, legal work, education and statewide initiatives. Its current support page reports more than 10,000 community members, 530 meetings annually and over 70 staff; these are scale claims, not health outcomes. The donation page explicitly identifies the recipient and EIN. Recipient and operations

The current refinery portfolio includes Long Beach storage expansion, Paramount/Altair conversion, national hydrogen-fluoride regulation and statewide fuel-policy litigation. Several cases involve Earthjustice, East Yard or other counsel; activity and shared representation do not identify a sole causal owner. Current cases

In August 2026 CBE launched a two-year, ten-organization statewide climate-adaptation cohort supported by the Robert Wood Johnson Foundation. That is current implementation capacity, but the announcement does not identify additional cooling/smoke protection, utilization or clinical effects. Existing foundation support belongs in the funding counterfactual. CACHE launch

2. Monitoring and information sharing

CBE's March 2026 letter recognizes progress on Exide cleanup while requesting better owner participation, retrospective sampling and state/federal contingency planning. It also identifies delayed technical-assistance grants and problems at other contaminated sites. These are plausible actionable bottlenecks, but the letter does not count additional homes cleaned because of CBE or report residents' health changes. Current intervention

The March legislative briefing says over 6,000 high-priority properties had been cleaned by end-2025 and describes remaining contaminated parcels and proposed new funding. It requests $70 million in each of two fiscal years, yet separately describes $70 million for about 1,000 parcels. That internal cost/scope ambiguity means neither $70,000 nor $140,000 is presented as a verified marginal cleanup price. The proposal is not treated as enacted funding. Primary budget record, p. 36

Useful monitoring would link otherwise-inaccessible households to actual access, completed remediation, sustained environmental and blood-lead changes, age-specific exposure duration and clinical outcomes. Measure what happens later without the additional advocacy, not only treatment versus no cleanup ever.

3. Qualitative assessment

The mechanism is biologically credible but heterogeneous. A Boston randomized trial of 152 young children with much higher baseline soil lead found roughly 1.5–1.9 micrograms/dL greater blood-lead declines from soil abatement than comparison interventions. Its median baseline soil concentration was 2,075 ppm; that is not directly portable to the lower-concentration Exide parcels now being considered. Original trial

The later HOME randomized trial enrolled 355 pregnancies and followed children through age eight. It reduced dust lead but found no statistically significant overall blood-lead reduction and no substantive neurobehavioral improvement; a subgroup had greater exposure reduction. This is not evidence that all remediation fails, but it makes a universal QALY-per-house assumption unsafe. Neither soil ppm nor an IQ/earnings estimate is silently converted to QALYs. Original HOME trial

The refinery air pathway is more directly modelable clinically: AQMD's draft assessment reports PM-related deaths and morbidity, rather than only tonnes or monetized benefits. However, Rule 1109.1 was adopted in 2021, with implementation scheduled over subsequent years. The historical total is not a new donor's impact. No current CBE-specific change to remaining compliance, associated emissions or LA/Orange exposure was established. Native table Adoption

Community voice, accountability and environmental justice can matter independently of health. Cleanup can also create dust, displacement or exported waste burdens. These are not presumed zero merely because clinical quantification is absent.

A more geographically analogous smelter-cleanup study in Omaha found lower elevated blood-lead odds in several models, but no intervention association in its fully adjusted difference-in-differences model. The authors caution that remediation and measurement timing were highly correlated. This strengthens the case for local longitudinal measurement rather than selecting only a favorable before/after result. Original researchers' summary

4. What do you get for your dollar?

Original returns report full recipient expenses of $4,803,145 (2022), $6,001,139 (2023) and $7,882,417 (2024); the mean is $6,228,900.33. No reported separately netted event/inventory costs require restoration. Original 2024 return

The central model charges five full budgets at the latest annual expense, not only a selected campaign cost. Technical/legal and organizing capacity changes the probability of additional enforceable refinery/fuel-related emissions controls beyond existing permits and Rule 1109.1. Current World Oil and Paramount disputes establish a live mechanism, not the assumed probability or PM reduction. HF accident prevention, soil lead, Bay Area activity and non-health justice value are not converted to PM mortality.

Conditional on an additional policy increment, 250,000 older residents experience 0.3 micrograms/m³ less annual ambient PM2.5. A 8% recipient-specific increase in that policy probability over the five-year effort and 35% marginal funding response are judgments, not measured attribution. The funding response allows for reserves and replacement gifts; coalition contribution is already inside the probability change and is not discounted twice. The assumed exposed cohort is alive at implementation, three years after model start.

Using the external HR 1.06 per 10 micrograms/m³, a 0.7 log-risk transport factor, annual baseline mortality hazard 0.035 and survivor utility 0.75, integrate competing all-cause survival curves for ten years and discount at 3%. Half the modeled effect is durable within that window; half is three-year acceleration of an otherwise identical policy. Acceleration retains only the finite survival difference versus delayed implementation. No lifetime QALYs are appended. All inputs beyond the external study estimate and financial accounts are judgments. External clinical study

The $10,000 normalization yields 0.0013250 local QALYs and $75,471,532 per 10. Favorable and poor joint scenarios give $516,735 and $3,795,077,413,953 per 10; they are sensitivity cases, not confidence limits. Zero and net-harm cases have no finite beneficial price. This is a partial-health estimate: omitted pathways have zero net central contribution by judgment, not proven absence of benefit or harm. Public implementation, regulated-facility, volunteer and partner resources are additional unpriced social costs; this is not an all-resource societal cost-effectiveness estimate.

Model, assumptions and sensitivity

Five full latest-year gross reported recipient budgets, across all programs and support, normalized linearly to a $10,000 ordinary gift. No policy-only cost carveout. External social resources unpriced.

Only health within Los Angeles and Orange counties (FIPS 06037, 06059) is counted. The 250,000 older adults are an analyst-selected conditional exposure cohort, not an office-count allocation or observed participant total. Riverside and San Bernardino outcomes are excluded. allPopulationQalys equals local quantified branch only, not an assumption that the whole portfolio is local.

h1=h0*exp(-ln(HR)/10*transport*pm); rho=ln(1.03); F(a,t)=(1-exp(-a*t))/a. qD=u*(F(h1+rho,T)-F(h0+rho,T)). a=min(A,T). qT=u*(F(h1+rho,a)-F(h0+rho,a)+(1-exp(-(h0-h1)*a))*(exp(-(h1+rho)*a)-exp(-(h1+rho)*T))/(h1+rho)). q=(f*qD+(1-f)*qT)/1.03^L. Qgift=10000/(E*Y)*b*(p*N*q-harm). Price=100000/Qgift if positive. N is cohort alive at implementation; no extra survival-to-implementation multiplier. Costs are constant-real full budgets; no cost discount haircut.

E
7882417 USD per recipient year (observed). Latest original filing, full reported gross expense. [cbe-99024]
Y
5 budget years (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
N
250000 local older people alive at implementation (judgment). Conditional policy exposure cohort selected for plausibility, not measured reach or demographic allocation.
Nall
250000 people in all quantified branches (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
pm
0.3 micrograms/m³ sustained ambient reduction (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
p
0.08 incremental policy probability (judgment). Low incremental probability over five years reflects multiple actors, existing rules and legal reversal; unelicited analyst judgment.
b
0.35 marginal funding response (judgment). Replacement funding/reserves uncertainty; smaller CBE response reflects materially larger reported unrestricted balances.
HR
1.06 hazard ratio per 10 micrograms/m³ higher PM2.5 (observed). External older US Medicare cohort estimate; conditional causal interpretation, not a CCA/CBE trial. [wu]
transport
0.7 multiplier on log HR (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
h0
0.035 annual all-cause mortality hazard (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
u
0.75 utility among survivors (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
T
10 finite years after implementation (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
L
3 years to implementation (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
f
0.5 durable branch share (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
A
3 years of acceleration (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.
harm
0 other QALYs lost per five-year effort before b (judgment). Explicit finite partial-health modeling judgment; requires empirical refinement.

Judgment central: partial older-adult air health: Cost: $10K; Los Angeles QALYs: 0.0013250029136660216; all-population QALYs: 0.0013250029136660216. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Favorable joint assumptions, not a confidence bound: Cost: $10K; Los Angeles QALYs: 0.1935226912898407; all-population QALYs: 0.1935226912898407. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Poor joint assumptions, not a confidence bound: Cost: $10K; Los Angeles QALYs: 2.6349923622729437e-8; all-population QALYs: 2.6349923622729437e-8. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

One-tenth central incremental policy probability: Cost: $10K; Los Angeles QALYs: 0.00013250029136660217; all-population QALYs: 0.00013250029136660217. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Strong replacement funding: Cost: $10K; Los Angeles QALYs: 0.0003785722610474348; all-population QALYs: 0.0003785722610474348. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Entire effect durable within finite ten years: Cost: $10K; Los Angeles QALYs: 0.0017169358830475157; all-population QALYs: 0.0017169358830475157. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Only three-year acceleration; finite survival comparison: Cost: $10K; Los Angeles QALYs: 0.0009330699442845277; all-population QALYs: 0.0009330699442845277. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Comparable three-year gross expense mean: Cost: $10K; Los Angeles QALYs: 0.0016767366521887274; all-population QALYs: 0.0016767366521887274. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

No causal clinical transfer: Cost: $10K; Los Angeles QALYs: 0; all-population QALYs: 0. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

No additional policy or health effect: Cost: $10K; Los Angeles QALYs: 0; all-population QALYs: 0. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

No policy benefit and ten QALYs of other harm per five-year effort: Cost: $10K; Los Angeles QALYs: -0.0008880524844092872; all-population QALYs: -0.0008880524844092872. All nonfinancial parameters except the external HR are explicitly analyst judgment; scenario bundle is not an empirical interval. Full five-year expense denominator; no gift restriction assumed.

Counterfactual: Already-completed properties, adopted air rules, public cleanup appropriations and work by regulators, residents and partner organizations continue without the marginal support. Acceleration must count only the exposure interval gained, not lifetime no-cleanup benefit. Technical/legal and organizing capacity changes the probability of additional enforceable refinery/fuel-related emissions controls beyond existing permits and Rule 1109.1. Current World Oil and Paramount disputes establish a live mechanism, not the assumed probability or PM reduction. HF accident prevention, soil lead, Bay Area activity and non-health justice value are not converted to PM mortality.

Attribution: p is the incremental recipient-specific probability effect of a five-year whole-organization effort versus partner/regulator action without it. b models donation-sensitive capacity after replacement funding, distinct from coalition p. Local linear extrapolation to a small gift is judgment and can fail at lumpy capacity thresholds.

Explicit judgment central, not identified causal charity effectiveness. Wide scenarios include clinical-null, no-policy-effect and harm. This is not a complete portfolio estimate or a reason to override donor diligence.

Sensitivity

  • Favorable/poor bundles vary several linked assumptions and are not probabilistic credible intervals.
  • Tenfold reduction in p increases price tenfold.
  • N and exposure are conditional on reform, not multiplied by total state population or total historic policy benefits.
  • Timing-only model compares gift and delayed-policy survival through the same finite endpoint.
  • Zero-net omitted pathways is a provisional central judgment; harms can reverse the result.

Unresolved inputs

  • Current otherwise-unfunded staffing/work plan and its effect on access, remediation completion or enforceable emissions controls.
  • Cleanup-counterfactual timing, local household age/exposure mix, post-cleanup blood lead and a defensible clinical/QALY dose-response with persistence.
  • Non-duplicated CBE contribution versus East Yard, residents, lawyers and agencies, plus marginal donor/public replacement.
  • Outcome-weighted LA+Orange share of statewide air, water, climate and other pathways; non-health value is not silently monetized.
  • Net cleanup/displacement/transport harms and external public/private implementation resource costs.

5. Funding and previous grants

The 2024 return reports revenue of $8,062,410, year-end cash of $5,989,690 and unrestricted net assets of $5,964,296. These balances do not prove that every asset is freely deployable today, but they weaken an assumption of immediate one-for-one expansion from an additional gift. Expense grew about 31% from 2023, while the return records 83 employees during the year; headcount is not full-time equivalents. Original return

A useful next-dollar case would specify otherwise-unfunded organizers or technical/legal capacity, the affected cleanup queue or enforceable emissions change, and why reserves, agency outreach, other funders or partner groups would not replace it. Current broad fundraising claims do not provide that marginal work plan. Support page The current jobs page lists no available positions; that is a website observation, not proof that the organization lacks useful work or cannot expand. Current vacancies

Annual expenses

Organization-level spending, including programs, administration and fundraising. The research list averages three comparable, consecutive full fiscal years when available.

  • FY 2022: $4.8M; Communities for a Better Environment, EIN 94-2998086, 12-month period, Accrual Form 990 full recipient expense; no reported netted event or inventory expense. Source
  • FY 2023: $6.0M; Communities for a Better Environment, EIN 94-2998086, 12-month period, Accrual Form 990 full recipient expense; no reported netted event or inventory expense. Source
  • FY 2024: $7.9M; Communities for a Better Environment, EIN 94-2998086, 12-month period, Accrual Form 990 full recipient expense; no reported netted event or inventory expense. Source

6. Sources

  1. 2024 original Form 990. Communities for a Better Environment / IRS. Published: 2025-11-18; retrieved: 2026-09-14.
  2. 2023 original Form 990. Communities for a Better Environment / IRS. Published: 2024-11-16; retrieved: 2026-09-14.
  3. 2022 original Form 990. Communities for a Better Environment / IRS. Published: 2023-11-15; retrieved: 2026-09-14.
  4. Current support page, recipient EIN and scale. Communities for a Better Environment. Published: not stated; retrieved: 2026-09-14.
  5. Comments on DTSC evaluation. Communities for a Better Environment. Published: 2026-03-09; retrieved: 2026-09-14.
  6. Current refinery litigation portfolio. Communities for a Better Environment. Published: not stated; retrieved: 2026-09-14.
  7. March 19 budget hearing, Issue 16. California Senate Committee on Budget and Fiscal Review. Published: 2026-03-19; retrieved: 2026-09-14.
  8. Exide residential cleanup and geography. California Department of Toxic Substances Control. Published: not stated; retrieved: 2026-09-14.
  9. Lead-contaminated soil abatement and urban children's blood lead levels. Weitzman et al. / JAMA. Published: 1993-04-07; retrieved: 2026-09-14.
  10. Residential lead-hazard intervention randomized trial. Braun et al. / JAMA Pediatrics. Published: 2018-08-06; retrieved: 2026-09-14.
  11. Refinery rule adoption and implementation horizon. South Coast AQMD. Published: not stated; retrieved: 2026-09-14.
  12. Draft socioeconomic assessment, native health incidence Table 27. South Coast AQMD. Published: 2021-09-07; retrieved: 2026-09-14.
  13. Soil remediation and children's blood lead at Omaha Superfund site. Ye et al. / US EPA. Published: 2022-04-04; retrieved: 2026-09-14.
  14. CACHE statewide adaptation cohort launch. Communities for a Better Environment. Published: 2026-08-14; retrieved: 2026-09-14.
  15. Current employment opportunities. Communities for a Better Environment. Published: not stated; retrieved: 2026-09-14.
  16. Evaluating long-term fine particulate exposure and mortality among the elderly. Wu and colleagues / Science Advances. Published: 2020-07-17; retrieved: 2026-09-14.