Concluding Comments

Because the PAM is an accounting framework composed of identities, users may choose among empirical techniques to estimate its elements. The budget-based approach described here has shortcomings. As this and subsequent chapters make clear, budgets place heavy reliance on judgment and informed guesswork by analysts. Statistical measures to indicate representativeness are often unavailable, output supply and input substitution responses to social prices are often incorporated in an approximate manner, disaggregations of input costs comprise many approximations, and fixed input-output coefficients are sometimes a necessary assumption. But budget-based approaches have great advantages as well. The data are relatively easy to collect and do not depend on the long time series that so often confounds econometric estimation.

Box 8.3. Decomposition of Nontradable Inputs for Wheat Production Systems in

This table contains 1981 data on intermediate inputs into a goodsoil wheat production system located in southern Portugal. The initial data are the costs per hectare for each intermediate input, shown as total private costs. The intermediate inputs are then classified as either tradables or nontradables. Three items-seeds, disinfectants, and fertilizer-are classified as tradables. The net policy transfer for fertilizer is a large subsidy (shown as 2,840.06) that reduces the private costs of farmers.

The other four intermediate inputs are classified as nontradable. The task is to decompose each of the four nontradable intermediates into tradable inputs, labor, and capital. Three principal distortions-a fuel tax, a labor tax, and a capital subsidyaffect the private prices of these nontradables. Only one distorting commodity policy, a 22 percent tax on fuel, affects the price and use of inputs into the nontradables. The private price of labor (market wage costs per hectare) is judged to be higher than it would be in the absence of policy because of legislation requiring vacation bonuses and social security contributions that do not have to be paid by other employers. Accordingly, all private labor costs exceed social labor charges by 20 percent. Government policy on capital is designed to create a subsidy that is enjoyed by wheat farmers and their input suppliers; the annual market interest rate is taken as 2 percent (in real terms, after correcting for inflation), compared with an estimated shadow real interest rate of 8 percent.

The net effect of these three kinds of policies on the four nontradables, shown in the right-hand column of the table, is a slight tax. When both tradable and nontradable intermediates are considered, the substantial subsidy on fertilizer ( 2,840.06) swamps the small net tax on nontradables (64.64) and creates a net policy transfer ( 2,775.42) that subsidizes 16 percent of the total social costs of intermediate inputs into wheat farming.

1The use of a disaggregated framework of costs simplifies the introduc-tion of more accurate information in a piecemeal fashion. Finally, the data can be selected to correspond to a number of research issues: the effects of policies on particular commodities, regions, or types of producers; the attractiveness of alternative technologies; or the effects of variations in access to input and output markets.