Knowledge that Transforms

Real exchange rates between the United States and its major trading partners were calculated for the Bretton Woods and flexible exchange rate periods. Unit root tests indicate that Purchasing Power Parity performed poorly in both periods. Tests for cointegration reveal limited instances in which it is possible to estimate the deviations from PPP as an error correcting model. The estimated error correcting models indicate that foreign, but not U.S., prices responded to deviations from PPP. Frenkel's (1981b) finding that Purchasing Power Parity (PPP) worked better during the 1920s than the 1970s caused considerable controversy. For example, Davutyan and Pippenger (1985) contend that the socalled collapse of PPP is a result of an increase in the relative importance of real versus monetary shocks. They argue that the 1970s, as opposed to the 1920s, was characterized by real supply shocks and the international coordination of monetary policies. The argument is that PPP did not fail; rather, there was an increase in the volatility of those factors giving rise to deviations from PPP. Hakkio (1984) reestimated PPP over the 1920s and 1970s; using cross-country tests (i.e., SURE estimates) to improve the efficiency of his estimates, he was able to support the hypothesis that PPP worked better in the 1970s than in the 1920s. On the other hand, papers by Adler and Lehman (1983), Dornbusch (1980), Frenkel (1981a), Junge (1985), and Krugman (1978) report findings contrary to the PPP hypothesis. Moreover, Kenen and Rodrik (1986) find that the volatility of real exchange rates has increased throughout the flexible rate period. This paper tries to shed some light on the importance and persistence of the observed deviations from Purchasing Power Parity under alternative exchange rate systems. While it is interesting to compare PPP in the 1920s versus the 1970s, it is equally useful to compare the 1960s versus the 1970s and 1980s. If real supply shocks and lack of monetary coordination are characteristic of the latter period, PPP should perform better in the 1960s. To illustrate the issues involved, consider the following econometric model of (Relative) Purchasing Power

Those with greater earnings capacity are likely to choose safer jobs, assuming safety is a normal good. Those who e xperience greater returns to job may choose riskier jobs. This paper estimates wage premia for risk of fatality and injury, allowing unobs ervables to affect earnings capacity and the returns to risk. As the endogeneity of job risk causes bias in OLS estimation, the model is e stimated with simultaneous equations and modified selection bias tech niques. The results indicate that unobserved heterogeneity in the ret urns to risk is important and that OLS underestimates the wage premia for fatality and injury risk.

Nonstationarity in the levels of spot exchange rates and domestic and foreign price indices makes the use of conventional tests of the absolute version of purchasing power parity (PPP) inappropriate. If PPP is true, inter-country commodity arbitrage ensures that deviations from a linear combination of spot exchange rates and domestic and foreign price levels should be stationary. Under these conditions, exchange rates and price levels should form a cointegrated system. We find the null hypothesis of no cointegration cannot be rejected for all five countries, thus violating the long-run absolute version of PPP.

The conventional paradigm that capital movements respond to differences in interest rates between countries and simultaneously reduce interest-rate differentials has been difficult to demonstrate empirically. This paper argues that such a demonstrati on may be feasible if a simultaneous model is specified that describe s the dynamics of adjustment and if a data interval is chosen that re veals the dynamics. Examination of U.S. and Canadian data from the 19 60s supports the argument-with monthly observations, that paradigm is strongly supported; with quarterly observations, capital flows and i nterest rates are not significantly related.

While voluntary energy conservation programs have been extensively promoted by electric utilities and public utility commissions, their effectiveness has been insufficiently critiqued. This paper contributes to the growing evaluation of such programs by measuring the net energy savings directly attributable to an actual set of programs. The analysis explicitly corrects for the self-selection bias that can arise in program evaluation. The correction is found to be important; traditional evaluation methods are subject to considerable bias. Correcting for this bias, the amount of program-induced energy savings is found to be considerably less than traditionally believed.

This paper estimates an empirical model of the choice between adjustable- and fixe d-rate mortgages using a large national database compiled by the Nat ional Association of Realtors. The explanatory variables in the probi t choice equation include borrower characteristics, regional dummies, the FRM rate, and the FRM-ARM rate differential. Since the intrerest rate on the mortgage not chosen is unobserved, this rate must be est imated in a way that eliminates selectivity bias. Simulation of the e stimated choice equation provides insight into the origin of the ARM revolution.

The effects of technological opportunity, market demand and RD with respect to productivity growth, it is not possible to distinguish demand and technological opportunity effects, but spillovers are important.