EFFECTS OF THE CPI BIAS ON INDIVIDUAL COHORTS
It is also interesting to ask which birth cohorts have benefited most from the housing bias.
From our earlier discussion, we expect the answer to be beneficiaries eligible under pre-1977
law. Under that regime, the initial benefit formula was price-indexed and all beneficiaries
would experience the full cumulative effect of the bias up to the time benefits terminate. Figure
4 presents our estimates by birth cohort of the present values of total benefit overpayments,
discounted to 1996 with the interest rate paid or projected to be paid on the OASDI Trust Fund.
Estimates are shown separately for pre-1977 and post-1977 Social Security law. Under the
former, overpayments grow as the birth year approaches 1916, reaching a peak of $18.6 billion.
Following the 1916 cohort, overpayments under pre-1977 law relate only to disability and
auxiliary beneficiaries and therefore drop sharply beginning with the 1917 cohort.
Nevertheless, all beneficiaries entitled under pre-1977 law who received benefits after 1974
were overpaid as a result of the CPI error.