The Office of the Chief Actuary (OCACT) is the actuarial arm of the Social Security Administration, responsible for producing cost estimates, demographic projections, and mortality tables for OASDI programs. In the context of DI research, OCACT is the primary — and for most periods, the only — source of systematic mortality data on DI disabled-worker beneficiaries.
Since at least the late 1970s, OCACT has published a numbered series of actuarial studies documenting DI disabled-worker mortality experience. Each study covers a multi-year observation window and produces period death probability tables by age, sex, and duration on the rolls (the Select and Ultimate Mortality Tables structure). Nine studies synthesized by Javier Meseguer (2021) span 1968–2015:
| Study No. | Period | Authors |
|---|---|---|
| 74 | 1968–1974 | Bayo and Wilkin (1977) |
| 75 | 1973–1976 | Bayo, Goss, Weissman (1978) |
| 81 | 1975–1978 | Schobel (1980) |
| 93 | 1977–1980 | Kelley and Lopez (1984) |
| 114 | 1991–1995 | Zayatz (1999) |
| 118 | 1996–2000 | Zayatz (2005) |
| 122 | 2001–2005 | Zayatz (2011) |
| 123 | 2006–2010 | Zayatz (2015) |
| 125 | 2011–2015 | Barrick-Funk (2020) |
Note the gap: no actuarial study covers the 1980s. This is a significant blind spot given the CDR crisis of that era (see Disability Insurance Chronology).
OCACT also produces the U.S. Social Security Area life tables (Bell and Miller, Actuarial Study No. 120, covering 1900–2100), which serve as the general-population benchmark in all DI mortality comparisons. OCACT's estimates differ from the Human Mortality Database (HMD): OCACT uses Medicare-enrolled population 65+ as its source for death counts and exposure, while HMD uses state-reported deaths and census estimates. The two sources produce meaningfully different estimates at older ages.
Separate from the DI actuarial study series, OCACT produces annual 75-year mortality forecasts as a key input to OASDI trust fund solvency projections. The method combines linear extrapolation of historical cause-specific death rates with subjective expert selection of 70 "ultimate rates of decline" (5 age groups × 2 sexes × 7 causes), which are approved by the OASDI trustees (political appointees). The reasoning behind specific choices is not publicly available. Soneji and King (2012) were the first to document this process in sufficient detail for replication and showed that a formal Bayesian alternative — incorporating demographic smoothness priors and smoking/obesity covariates — outperforms OACT's method in out-of-sample validation and implies substantially worse solvency: trust fund balance $730 billion lower and negative net income 2–3 years earlier by 2031. See SSA Mortality Forecasting.