Academic journal article International Advances in Economic Research

Regional Tests of Okun's Law

Academic journal article International Advances in Economic Research

Regional Tests of Okun's Law

Article excerpt


"Okun's law" describes an enduring empirical observation first made by Arthur Okun, relating departures from the natural rate of unemployment to changes in real output. This paper employs a recent development in trend-cycle decomposition of economic time series to measure the Okun coefficient using U.S. national and regional data. The key finding of this paper is that the value of the coefficient measuring the change in real output per unit of change in the unemployment rate, both measured as departures from equilibrium, is stable at a value of about 2 for all time periods and across regions of the U.S., irrespective of the method used to measure equilibrium output and employment. (JEL E32)


Arthur Okun [1962, p. 99] observed that "in the postwar period, on the average, each extra percentage point in the unemployment rate above four percent [that is, his estimate of what we would now refer to as the natural rate of unemployment] has been associated with about a three percent decrement in real GNP." Because Okun's empirical finding held up well during the ensuing decade, the 3:1 trade-off between real gross national product (GNP) growth and the unemployment rate became known as Okun's law. Since that time, a number of papers have established theoretical foundations for Okun's law [Clark, 1983; Gordon, 1984; Prachowny, 1993] and tested the stability of the 3:1 tradeoff [Clark, 1983; Gordon, 1984; Adams and Coe, 1989; Holloway, 1989; Prachowny, 1993; Attfield and Silverstone, 1998]. In general, Okun's law has withstood most challenges, although current estimates of the trade-off fall into a range closer to 2:1 than 3:1 [Gordon, 1984; Attfield and Silverstone, 1998; Moosa, 1997] and vary according t o the methods and specifications used. Variations notwithstanding, the stability of Okun's law contrasts favorably with the Phillips curve, its counterpart in the unemployment-inflation space. [1]

Because of the way that Okun's law is framed, tests of its proposition involve two imposing empirical problems. First, the law refers to departures of real gross domestic product (GDP) [2] and unemployment from their long-run or equilibrium trends, with the latter often referred to as potential GDP and the natural rate of unemployment, respectively. These equilibrium trends must be estimated either by time series methods [Moosa, 1997] or by construction using a modified production function approach in which potential output is estimated using fully employed inputs [Gordon, 1984; Adams and Coe, 1989]. Either approach invites considerable criticism of assumptions and methods, [3] and there is likely to remain substantial measurement error in any computation of potential or long-run quantities. Further, the latest year of observation for these constructed data sets is 1988 when the natural unemployment rate was estimated by Gordon [1997] to be in excess of 6.0 percent, well above the levels we have experienced with stable to declining inflation in the mid to late 1990s.

Second, in formulating his law, Okun clearly seems to be referring not to a ceteris paribus relationship between changes in the unemployment rate and changes in real GDP. Writing later, Okun [1970, p. 140] assumed that other factors and inputs would be changing pari passu with employment: "The 3 percent result [from a reduction of 1 percent in the unemployment rate] implies that considerable output gains in a period of rising utilization rates stem from some or all of the following: induced increases in the size of the labor force; longer average weekly hours; and greater productivity." Thus, an appropriate specification of the effect of changing unemployment rates on output would include factors such as capital inputs, labor hours, and participation rates, all measured as deviations from long-run trends. Because the levels of these variables are extremely difficult to measure, let alone their potential values, attempts to include them in tests of Okun's law have produced mixed results [Prachowny, 1993; Attf ield and Silverstone, 1997]. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.