DEVELOPMENTS WITHIN THE ECONOMY and the profession in recent years have generated a marked swing toward pessimism in the appraisal of the tradeoff between inflation and unemployment. During the fifties and the sixties, a 4 percent unemployment rate was generally accepted as a target for full employment. At that time, it was expected that it might be accompanied by an inflation rate of 2 or, at most, 3 percent.' In contrast, some recent readings of the Phillips curve reported in this journal suggest that, under present circumstances, a 4 percent unemployment rate means as much as 5 percent inflation for the long run, while holding the inflation rate down to 2 percent would require an unemployment rate of 51/2 percent.2 Some economists have responded to this unhappy news by espousing either "brake riding" or "gas pumping," as I have characterized these positions elsewhere.3 The brake rider would accept a higher unemployment rate. Although some brake riders have urged actions to ameliorate the adverse human and redistributive impact of unemployment, their program still involves large 1. See, for example, the estimates in Paul A. Samuelson and Robert M. Solow, "Analytical Aspects of Anti-inflation Policy," in American Economic Association, Papers and

Upward Mobility i n a High-pressure Economy THE CHOICE OF AN AGGREGATE TARGET of resource utilization remains one of the key issues facing policy makers and macroeconomists. Obviously, fuller utilization of labor and capital brings benefits in the form of extra incomes, output, and jobs; at the same time, it clearly imposes costs by increasing inflationary tendencies. Various economists see these benefits and costs very differently. Henry Wallich once suggested that macroeconomists could be classified into advocates of "high pressure" and "low pressure" operation of the economy.' At the present time, the controversial range for the target unemployment rate extends from 4 to 5 percent. Generally, high-pressure advocates concede that, with existing labor market institutions, unemployment rates below 4 percent would be associated with unacceptable inflation; while most low-pressure advocates agree that un-One part of the gap calculation is merely a piece of arithmetic. If participation rates, average hours, and manhour productivity are held constant, the extra jobs associated with reducing the unemployment rate from 5.0 to 4.0 percent would add 1/95 to total labor input and, by assumption, that much-1.05 percent-to real output. Various researchers have obtained similar results in estimating the lengthened workweek-increased overtime and diminished part-timeassociated with reduced unemployment. Black and Russell, Thurow and Taylor, Perry, and I, have all found that the increment in total manhours due to the lengthening of the workweek is between 0.35 and 0.5 percent for a drop of 1 percentage point in the unemployment rate.3 Taylor and Thurow explicitly recognize the use of overtime as a temporary buffer in periods of rapid growth of output; holding the growth of output constant, they estimate that the workweek expands by 0.4 percent for a 1 percentage decrement in the unemployment rate. Perry's estimate standardizes for differences in average hours among various demographic groups; he finds an increase of 0.35 percent in average hours for a decline of 1 point in the official unemployment rate. I grasped originally at the straw of a calculation of normal labor force by the Bureau of Labor Statistics and guessed that there might be three "hidden unemployed"-people not seeking work actively who would take jobs in a high-utilization economy-for every ten recorded unemployed in excess of a 4 percent unemployment rate. Other studies have generally produced much larger estimates of the number of hidden unemployed. But some of the largest estimates of the participation response are based on statistical techniques that may create an upward bias, as Jacob Mincer demonstrated.4 Perry's recent study disaggregates the demographic groups 3.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Brookings Institution Press is collaborating with JSTOR to digitize, preserve and extend access to Brookings Papers on Economic Activity.

State and Local Budgets t he Day after I t Rained: Why I s t he Surplus So High? READERS of the financial press will be shocked to find that, in the aggregate, the 78,000 state and local governments in this country are running a hefty budget surplus. These governments had a combined budget surplus of $29.2 billion in 1977, by far the largest ever recorded, and in part offsetting the impact of the one government most noticeably not in surplus (the federal deficit was $49.6 billion in that year). The rise in the state and local surplus has been exceedingly dramatic: at its recession low in the first quarter of 1975 it was $3.7 billion; by the third quarter of 1977 it had risen to $32.9 billion, accounting for over one-third of the national rise in gross saving over this two-year period. Such large changes in the saving behavior of any sector are interesting in their own right, and they have important macroeconomic implications for short-run stabilization policies and long-run growth. In this report I examine both matters. I first disaggregate the budgetary numbers to explore separately movements in the surplus of state and local pension funds and general governments on both current and capital accounts. I then estimate some time-series equations explaining state and local budgetary magnitudes up to 1974, making out-of-sample extrapolations of these Note: Mark Greene has helped me understand the Michigan computer, John Gorman the treatment of pension fund surpluses, Serge Taylor environmental impact statements, and Roger Vaughan the local public works bill. I would also like to thank Ronald Ehrenberg, Alan Fechter, Harvey Galper, Robert Hartman, Robert Reischauer, and Daniel Rubinfeld for their comments on an earlier version.

Can t he Inflalon of t he 1970sbe Explained?MANY STANDARDS inflation has been a "surprise" during the past six years. Errors in forecasting inflation have increased markedly compared with earlier periods. For instance, during the interval 1971:3 to 1975:4 the root mean-square error of the Livingston panel of economists in forecasting the consumer price index six months ahead was 3.5 percentage points at an annual rate, compared with an error of 1.6 percentage points over the previous seventeen years.' Not only did the panel forecasters fail to predict the increased variance of the inflation rate in the 1970s, but also they fell far short in predicting the cumulative total price change between 1971 and 1976-24.0 percent compared with the actual change of 34.0 percent.2 Most of the error occurred during the four quarters of 1974, Note: This research has been supported by the National Science Foundation. I am grateful to my research assistant, Joseph Peek, for his superb efficiency in compiling and creating the complex data base on which the paper depends. Helpful suggestions were received from participants in seminars at Northwestern, the University of California at Berkeley, and the Federal Reserve Banks of San Francisco and Philadelphia. 1. The Livingston forecasts were obtained from John A. Carlson, "A Study of Price Forecasts," Annals of Economic and Social Measurement, vol. 6 (Winter 1977), table 1, pp. 33-34. I calculated the errors by comparing the six-month-ahead forecasts with the change in the consumer price index in the two relevant quarters. For instance, Carlson's calculation of the predicted quarterly rate of change between December 1973 and June 1974 is compared with the average quarterly rate of change of the CPI in the first and second quarters of 1974. The "previous seventeen years" runs from 1954:1 to 1971:2. 2. The actual figure refers to the sum of the quarterly rates of change of the CPI in the interval 1971:3 through 1976:2. The forecast figure is the sum of the sixmonth predicted changes calculated by Carlson from the Livingston panel data for the ten surveys between June 1971 and December 1975.253 3. The errors for the forecasts from five large-scale models compiled by McNees were similar. The four-quarter-ahead forecast made in 1973:4 for the change in the GNP deflator to 1974:4 was 6.