Inflationary Shocks
Nominal contracts

One of the greatest economic puzzles in an age of widely varying, random rates of inflation is the persistent use of nominal contracts promises of future payment of prespecified, non-contingent sums of fiat money. In Discussion Paper No. 602, Scott Freeman and Guido Tabellini look for reasons why such contracts may be optimal for a large class of environments. Starting from the observation that each individual is generally a party to several contracts with several others, the authors develop an overlapping generations, general equilibrium model in which the equilibrium contract form is optimal given the contracts elsewhere in the economy. They show that there are economic environments in which nominal contracts may be superior to those contingent on either relative or aggregate price level shocks.
Freeman and Tabellini show that nominal contracts are optimal under aggregate shocks in two cases. First, if all contracts are nominal, the contracting parties share aggregate risk in proportion to their net wealth. If all individuals have the same constant degree of relative risk aversion, this is exactly what optimal risk sharing requires.
Second, if individuals differ in the extent of their risk aversion, there are gains from having contracts contingent on aggregate shocks; but such gains are generally of second-order magnitude. Wealthier individuals therefore bear more risk, though not necessarily in proportion to their wealth, and the small costs of incorporating contingencies may therefore restore the optimality of fixed nominal contracts.
When contracts are payable in fiat money, contingencies that reduce the uncertainty of final real wealth generally increase the uncertainty of cash flows; so contingent contracts payable in fiat money will always entail the cost of holding enough cash to meet the maximum contingent payment specified by the contract. Freeman and Tabellini show that fixed nominal contracts may be optimal even if aggregate shocks are observable at zero cost and individuals differ in their attitudes towards risk. Also, nominal (non-contingent) contracts remain optimal even in the presence of relative price shocks, provided that the contracting parties are ex ante identical, i.e. subject to the same uncertainties concerning their future preferences.
Finally, even when nominal contracts are not optimal in the presence of either relative or aggregate risk, there is a combination of contingent and nominal contracts that turns out to be optimal: such contracts may be interpreted either as equity or future contracts or explicit insurance.

The Optimality of Nominal Contracts
Scott Freeman and Guido Tabellini

Discussion Paper No. 602, October 1991 (IM)