Calling the One-Sided Bet: A Case Study of Budget Scoring in the 1996 Farm Bill

During the 1995–96 budget reconciliation/farm bill debate, analysts at the Congressional Budget Office (CBO) developed systematic techniques for probability scoring the estimated costs to the federal budget of changing commodity loan programs. Probability scoring is used to address onesided bets. In a legislative context, a one-sided bet is a legislated change in program operation (usually a program parameter) that does not cause a change in estimated government costs when those costs are estimated against the baseline projection path, but could cause a change in government costs in one direction only if future developments differ from assumptions in the baseline. With the use of probability scoring, the CBO estimated that farm bill proposals that raised or lowered crop loan rates correspondingly raised or lowered loan program costs. These results did not occur under standard point-estimate scoring because baseline point estimates of projected prices were too far above crop loan rates to show much change in loan program costs when loan rates changed. In the final 1996 farm bill as enacted, Congress capped future loan rates for grains and cotton. This is the first time such caps have been imposed since formulas to set loan rates based on moving average prices were introduced in 1986. Congressional staff have indicated to the CBO that without the probability-scored savings as an incentive, the Congress probably would not have capped crop loan rates. Indeed, the staff indicated that without probability-scored costs as a disincentive, the Congress might well have raised crop loan rates.