The 1920-21 deflation: the role of aggregate supply.
The 1920-21 recession in the United States was brief relative to the Great Depression of a decade later, but it included a remarkably sharp price deflation. The decline in the GNP price deflator from 1920 to 1921 is the largest one-year percentage decline in the series in the more than 120 years covered. This is true whether the Department of Commerce  estimates or the recently provided Balke and Gordon  or Romer  estimates are used. These estimates produce one-year deflation figures of 18 percent, 13.0 percent, and 14.8 percent, respectively. The closest competitor is the 11.5 percent deflation recorded for 1931-32, the third year of the Great Depression. (1)
Annual data for wholesale prices tell a similar story. Wholesale prices declined by 36.8 percent for 1920-21, the largest one-year decline on record, going back at least to the American Revolutionary War period.
The 1920-21 deflation contains another striking feature. Not only was it sharp, it was large relative to the accompanying decline in real product. The ratio of the percentage decline in the GNP deflator for 1920-21 to the percentage decline in real GNP is 2.6 using the Department of Commerce figures, 3.7 using the Balke and Gordon data, and 6.3 using the Romer data. By contrast, during 1929-30, the first year of the Great Depression, the GNP deflator declined by 2.7 percent and real GNP by 9.4 percent, for a ratio of 0.3. The ratios of the percentage decline in GNP prices to the percentage decline in real GNP for 1930-31, 1931-32, 1932-33, and 1937-38, the other Great Depression years in which real GNP declined, were 1.0, 0.9, 1.2, and 0.3, respectively, all well below the 1920-21 figures.
This paper examines the 1920-21 deflation. It asks why the deflation was so sharp, both in itself and in relation to the decline in real product. The answer, the paper concludes, is that the deflation was produced by a sharp decline in aggregate demand combined with an increase in aggregate supply, a supply increase in which deflationary expectations played a prominent role.
II. THE 1920-1921 RECESSION
The National Bureau of Economic Research dates the 1920-21 recession from a general business peak in January 1920 to a trough in July 1921. It was mild at first. Wholesale prices continued to increase until May 1920, four months past the general business peak. By July 1920, the Federal Reserve Board's index of industrial production had declined by only 7 percent from its January peak, and factory employment had fallen 7.3 percent.
The contraction then became severe. By the year's end, industrial production had fallen 25.6 percent below its January 1920 peak and bottomed out at 32.6 percent below its January 1920 level in July 1921, the general business trough. Wholesale prices were 42.9 percent below their May 1920 peak by July 1921. Industrial production had fallen by 32.6 percent in eighteen months, wholesale prices by 42.9 percent in fourteen months. The deflation eliminated more than 70 percent of the rise in wholesale prices associated with World War I.
Civilian unemployment rose substantially during the recession. According to Lebergott [1964, 512], the unemployment rate was 1.4 percent for both 1918 and 1919, 5.2 percent for 1920, and 11.7 percent for 1921.
The 1920-21 recession was not the first following the end of World War I. The economy had slowed in 1918, at about the time of the November armistice, as the transition was made from wartime to peacetime production. Both industrial production and wholesale prices declined moderately. The National Bureau of Economic Research dates this first post-war recession as extending from August 1918 to March 1919.
The economy then turned upward and expanded, peaking in January 1920. In this period industrial production rose by 25.5 percent. Wholesale prices, after hitting bottom in February 1919, had risen by 20.7 percent by January 1920, the general business peak, and by 28.0 percent at their own May 1920 peak. (2)
III. THE DEMAND VIEW
Friedman and Schwartz [1963, 205-39] attribute the severe phase of the 1920-21 recession and its attending deflation to monetary restraint. Monetary policy was expansive throughout World War I, including the period of U.S. neutrality. Policy remained expansive during most of 1919, even though by summer an inflationary boom was underway. The Federal Reserve was pegging interest rates at a low level using its loan discount rate in order to accommodate the Treasury's funding of the war debt. The Fed also had an interest in protecting commercial bank portfolios, which contained substantial quantities of war bonds and loans secured by war bonds.
Monetary policy began to shift in December 1919, then changed markedly in January 1920. The Federal Reserve Bank of New York's discount rate, which had been pegged at 4 percent since April 1919, was raised to 4.75 percent in December 1919, to 6 percent in January 1920, and to 7 percent in June 1920. Similar discount rate increases were made at the other Federal Reserve Banks.
Friedman and Schwartz argue that these sharp increases came too late to be responsible for the January 1920 turning point but that they produced the severe contraction and deflation which came after mid-year. The monetary base and the M2 money supply merely leveled off during the first half of 1920, as the higher discounting costs had only a moderate effect on commercial bank discounts at Federal Reserve banks for a time. However, both the monetary base and the M2 money supply peaked in the autumn and then declined substantially over the next year. The decline in the monetary base from October 1920 through January 1922 was the largest recorded for so short a period in the Friedman and Schwartz series, a series dating back to 1867. The M1 money supply peaked earlier, in March 1920, but Friedman and Schwartz emphasize the behaviors of the M2 money supply and the monetary base in their discussion. (3)
Wilson [1948, 93-114] offers a more Keynesian explanation for the 1920-21 recession and deflation. In Wilson's view, the primary factors were a shift in the Federal budget from a deficit to a surplus position and a decline in net commodity exports, the very factors whose strength had fueled the preceding business expansion. Tighter monetary policy induced increases in interest rates, and these were a contributory factor in the recession, Wilson asserts, but they were not a main factor.
Gordon's [1974, 17-23] explanation for the deflation is similar to Wilson's in that he points to the emergence of a Federal budgetary surplus and a decline in exports. But Gordon attaches more importance to monetary restraint, assessing it as almost of equal importance. He also cites post-war expectations of price and sales declines as a factor depressing business purchases and consumer resistance to high prices as still another factor.
Roose  suggests endogenous supply bottlenecks were a factor in the January 1920 downturn, applying a theoretical argument credited to Hicks [1950, 128-35]. In this argument, production ceilings check the growth rate of consumption goods production and ultimately produce a decline in investment expenditures through an accelerator effect. Multiplier effects on consumption expenditures follow.
Roose is not willing to state these supply bottlenecks and their induced aggregate demand effects were the primary cause of either the upper turning point or the subsequent contraction in output and an autonomous decline in aggregate demand, identifying the shift to monetary restraint, the shift in the Federal budget from deficit to surplus, the decline in commodity net exports, and a mid-1919 decline in construction as factors exerting deflationary pressures on the economy (see Roose [1958, 350]).
Pilgrim , like Friedman and Schwartz, Wilson, Gordon, and Roose, sees a decline in aggregate demand as the primary factor in the severe phase of the contraction. He considers specifically whether it was monetary restraint or a decline in autonomous expenditures, especially the shift to surplus in the Federal Government's budget position, that was the major source of the fall in aggregate demand. He concludes that monetary restraint played the larger role, arguing that tighter monetary policy induced increases in the cost and availability of credit which produced the contraction by discouraging investment expenditures, first in construction spending, and later in fixed investment at large.
Pilgrim makes the significant point that supply factors played a role in the timing of the upper turning point in January 1920. He sees the five-week bituminous coal strike commencing on November 1, 1919 as being especially important. The strike was brief and coal production was restored promptly, but a government order restricting supplies to manufacturing in favor of other coal consumers aggravated the effects on manufacturing output and extended them into 1920. A brief railroad strike in April 1920 capped the difficulty, causing some plants to reduce or shut down production temporarily. Such effects are consistent with the rise in wholesale prices lasting until May 1920, well past the general business peak in January. However, Pilgrim concludes that the effects of the strikes were too short-lived to have caused the severe phase of the contraction, which appeared only after mid-year.
IV. THE ROLE OF AGGREGATE SUPPLY IN THE DEFLATION
The explanations for the 1920-21 deflation reviewed above agree on one important point. Prices declined in response to a decline in aggregate demand. A puzzle remains, however: Why was the deflation so large relative to the decline in real product, compared to the experience of the Great Depression, barely a decade later?
It is tempting to cite the sharp increase in the civilian labor force for 1920 over 1919 (Table I) as a substantive factor, one leading to a rightward shift in the short-term aggregate supply of output curve. Civilian money wages and production costs would have been especially sensitive to this labor force measure.
TABLE I The Total Labor Force, The Armed Forces, and The Civilian Labor Force, 1914-22 (a) Total Labor Force [DELTA] %[DELTA] 1914 39,564 1915 39,774 210 0.50% 1916 40,238 464 1.20% 1917 40,742 504 1.30% 1918 41,980 1,238 3.00% 1919 41,239 -741 -1.80% 1920 41,720 481 1.20% 1921 42,341 621 1.50% 1922 42,772 431 1.00% Armed Force [DELTA] %[DELTA] 1914 163 1915 174 11 6.70% 1916 181 7 4.00% 1917 719 538 297.20% 1918 2,904 2,185 303.90% 1919 1,543 -1,361 -46.90% 1920 380 -1,163 -75.40% 1921 362 -18 -4.70% 1922 276 -86 -23.80% Civilian Labor Force [DELTA] %[DELTA] 1914 39,401 1915 39,600 199 0.50% 1916 40,057 457 1.20% 1917 40,023 -34 -0.10% 1918 39,076 -947 -2.40% 1919 39,696 620 1.60% 1920 41,340 1,644 4.10% 1921 41,979 639 1.50% 1922 42,496 517 1.20% (a) From Lebergott (1964, 512, Table A-3). The data are constructed so as to be comparable to averages of monthly values.
The 4.1 percent increase in the civilian labor force for 1920 is its largest one-year increase in the entire period (1900-90) covered by the combined Lebergott [1964, 512] and Bureau of Labor Statistics (U.S. Council of Economic Advisers [1991, 322]) data, excluding only 1946 (6.8 percent) and 1947 (4.6 percent), the years following the end of World War II, when almost 10 million service personnel were demobilized. A Chow  test fitting the percent increase in the civilian labor force to the real wage and a trend variable, using annual data for 1901-26, (4) first excluding 1920, and then including it, indicates that the 4.1 percent figure does reflect a statistically significant shift in the relationship.
Demobilization of service personnel was the most important factor in the 1920 civilian labor force increase, though a rise in immigration also contributed. The armed forces declined by 1,361,000 for 1919 and 1,163,000 for 1920. (5) Immigration, which had been interrupted by the World War I hostilities, rose from 110,618 for 1918 and 141,132 for 1919 to 430,001 for 1920 and 805,228 for 1921 (U.S. Department of Commerce [1960, C 88], for years ending June 30).
However, the timing of the demobilization of service personnel in relation to the wholesale price peak suggests that the 1920 increase in the civilian labor force was not a major factor in the 1920-21 deflation. The 1919 and 1920 civilian labor force figures are mid-year figures, reflecting the increase at mid-year 1920 over mid-year 1919, with no monthly data available. Demobilization was largely complete by the beginning of 1920 (Mock and Thurber [1944, 134-35]), too early to have contributed substantially to a deflation beginning in May 19206 unless lags between demobilization and entry into the civilian labor force were fairly long. (7)
A second possibility is that the short-term supply of aggregate output curve was still fairly steep in 1920, as compared to 1929, because price and wage stickiness was less pervasive in 1920. Basic factors creating a long-term trend towards price and wage stickiness--urbanization, industrialization, unionization--had had less time to do their work. However, the data do not substantiate this explanation very well. Table II reveals that the 1920-21 ratio of the percentage change in the GNP deflator to the percentage change in real GNP was also larger than those for the three pre-World War I twentieth century recessions. Several exceptions occur during the nineteenth century, for more modest deflations.
TABLE II Ratios of Percentage Change in GNP Deflator to Percentage Change in GNP and Components: Recession Years, 1893-1921 Recession Year Kendrick Series (a) Recession % [DELTA]P/ Year % [DELTA]Y % [DELTA]P % [DELTA]Y 1893-94 2.3 -6.3 -2.7 1895-96 1.2 -2.5 -2.1 1903-04 -1.0 1.3 -1.3 1907-08 0.1 -0.7 -8.2 1913-14 -0.1 1.0 -7.7 1920-21 6.4 -14.8 -2.3 Department of Commerce Series (b) Recession % [DELTA]P/ Year % [DELTA]Y % [DELTA]P % [DELTA]Y 1893-94 1895-96 1903-04 1907-08 1913-14 -0.6 2.3 -3.9 1920-21 2.6 -18.0 -7.0 Recession Balke and Gordon Series (c) Year % [DELTA]P/ % [DELTA]Y % [DELTA]P % [DELTA]Y 1893-94 1.8 -5.3 -2.9 1895-96 -0.1 0.3 -2.3 1903-04 0.0 0.1 3.8 1907-08 0.4 -2.1 -5.5 1913-14 -0.2 1.3 -7.6 1920-21 3.7 -13.0 -3.5 Romer Series (d) Recession % [DELTA]P/ Year % [DELTA]Y % [DELTA]P % [DELTA]Y 6.9 -6.2 -0.9 1893-94 -1.0 -2.6 2.7 1895-96 3.7 1.1 0.3 1903-04 0.2 -0.7 -4.2 1907-08 0.4 1.0 -2.3 1913-14 6.3 -14.8 -2.4 1920-21 (a) U.S. Department of Commerce , Series Al and A7. (b) U.S. Department of Commerce , Tables 1.2, 1.25, and 7.4. (c) Balke and Gordon [1989, 84-85]. (d) Romer [1989, 22-23].
A third possibility is that the disturbance provoking the 1920-21 deflation may have been concentrated more than usual in the agricultural sector, a sector historically subject to relatively wide price changes. Soule [1947, 99-100] suggests this possibility, pointing to the increase in the supply of foreign agricultural commodities that occurred following World War I as European production recovered from its wartime disruptions and shipping became available to transport surpluses which had accumulated in other regions. (See also Romer [1988, 110-11] on this point.)
Strictly speaking, this would be an aggregate demand disturbance insofar as U.S. real GNP and GNP prices are concerned. It consists of a decline in the net exports component of aggregate demand, the same disturbance noted by several of the writers discussed above. In this case both foreign and domestic demands for domestic agricultural product declined, representing a shift from domestic to foreign product. However, there can also be an effect on the domestic aggregate supply curve, again on the slope. If the supply of domestic agricultural product is less price elastic than the supply of other domestic product, the disturbance steepens the domestic aggregate supply curve as compared to a demand disturbance which is neutral between agricultural and non-agricultural product. This aggravates the decline in the GNP price deflator, both absolutely and relative to the decline in real GNP.
It does not seem likely that this factor was strong enough to in itself to account for the sharpness of the 1920-21 price deflation. If wholesale prices excluding farm products and food are substituted for GNP prices in the ratios of the percentage decline in prices to the percentage decline in GNP presented in the introduction, the ratio for 1920-21 (which is 5.0) is still larger than those for 1929-30, 1930-31, 1931-32, 1932-33, and 1937-38 (which are 0.7, 1.4, 0.5, -0.5, and 0.9, respectively), and for 1913-14 (which is 1.3) as well. (8) Obviously, the sharper decline in the prices relative to the decline in output in 1920-21 extended beyond food and farm products. Of course, it can be argued that food and farm products are significant intermediate products for some of the nonagricultural price categories. Still, domestic product originating in the farm sector was only 13 percent of gross domestic product in 1920 (U.S. Department of Commerce [1960, F 44-46]).
A supplementary explanation, and a persuasive one, is that the aggregate supply curve was increased during 1920 by deflationary expectations. Prices had run up sharply during 1914-20, increasing by 115 percent, according to Department of Commerce estimates. Previous major wars had been accompanied by large inflations and followed by sharp deflations, facts which were well known from the history books if not firsthand.
Klein  has offered a more broadly based theory of price expectations which fits the 1920-21 deflation very well. Klein argues that "gold standard expectations" prevailed until the 1960s, when the reality of a fiat money supply finally was perceived. With gold standard expectations people believe that large price changes in one direction will be reversed, since prices are tied to money and money is thought to be tied to gold. According to this theory, the large 1914-20 inflation, capping an inflation actually beginning in 1896, created postwar deflationary expectations, increasing the aggregate supply curve. When aggregate demand declined in 1920, the price decline was aggravated and the output decline cushioned. By contrast, when aggregate demand declined in 192933, prices had been relatively stable for six years, so that deflationary expectations were not nearly so pronounced.
The return of labor peace by mid-1920 is still another supply-side factor. The year 1919 was one of unusual labor force strife, with effects extending into early 1920. Almost 4 million workers were on strike at one time or another during 1919 according to estimates by Bing [1921, 293]. (9) This compares to about 1.2 million each for 1917 and 1918. Bureau of Labor Statistics data, which begin with 1927 (see U.S. Department of Commerce [1960, D 765]), record no figure larger than 1919's 4 million, excepting only the 4.6 million figure for 1946. Pilgrim [1974, 280-82] lists strikes in key industries as occurring in September 1919 (iron and steel), November 1919 (bituminous coal), and April 1920 (railroads). As noted above, the bituminous coal strike had far reaching and long lasting effects on other industries. Just as these strike effects restricted output in late 1919 and early 1920, aggravating inflation and helping to prolong the rise in wholesale prices until May 1920, several months past the general business peak, the return of labor peace following the April 1920 railroad strike aggravated the subsequent price deflation and cushioned the decline in aggregate output.
The unusual labor strife of 1919 was largely a product of labor's efforts to consolidate its wartime gains in the face of a deteriorated market position. The gains were very real. Average annual earnings in manufacturing had increased by 51 percent during 1916-18, while consumer prices had increased by only 38 percent. And the higher real annual earnings were obtained in a slightly shorter work week, due mainly to the Federal Government's requirement of an eight-hour regular work day on war contract work (see U.S. Department of Commerce [1960, D 589, D 605, E 113] and Hughes [1977, 138-43]).
These gains resulted from labor's strong wartime market position. The demand for goods and services rose sharply, while the supply of labor to produce them was constrained by the military's manpower needs and the low level of immigration during the war years. Augmenting labor's position was the government's interest in uninterrupted production, and its greater coercive powers over firms than workers, which led it to be generally supportive of labor in labor-management disputes.
Once the armistice was signed, labor's position deteriorated rapidly. Government war contracts were cancelled almost immediately, shrinking the area where gains had been greatest. War boards and agencies were dismantled promptly, and the government's role in labor-management disputes became one of conciliation. Labor strife rose and labor's success declined. According to one survey, of the settlements clearly favoring either management or labor, the percentage favoring labor fell from 58.6 percent in 1918 to 46.1 percent in 1919 (Bing [1921, 296]). By the spring of 1920, with unemployment rates rising, labor ceased its aggressive stance and labor peace returned.
The 1920-21 deflation is one of the sharpest in U.S. history, the largest one-year deflation going back at least to the American Revolutionary War. Not only was the deflation large, it was large relative to the accompanying decline in real product.
A common view has been that the deflation resulted from a decline in aggregate demand. A better explanation is that it resulted from a decline in aggregate demand combined with an increase in aggregate supply. Deflationary expectations were one important supply-side factor. The retreat of labor strife was another.
It is natural to ask why a 1920-21 type deflation did not follow World War II. The first year after the war, 1946, was characterized by a high degree of labor strife, just as was 1919, and the labor strife subsequently declined. Demobilization of service personnel had greater potential as a deflationary factor following World War II, since the World War II demobilization was larger, even allowing for a larger population and labor force. Moreover, World War II was accompanied by a large inflation (though not so large as the inflation accompanying World War I). The GNP deflator rose by 53 percent from 1936 to 1946. Klein [1975, 472] argues that gold standard deflationary expectations were present following World War II and cites the Joseph A. Livingston survey of price anticipations data to document their presence. (10) Yet, despite these deflationary supply-side factors, annual data for the GNP price deflator record only a single decline for the decade following World War II, a decline of less than one-half percentage point for 1948-49.
The reason for the absence of a 1920-21 type deflation following World War II is clear. Such a deflation could not occur following World War II, even assuming similar supply conditions (including no increase in price and wage stickiness), because there was no sharp decrease in aggregate demand. Fueled by strong demands for investment goods and consumer durable goods backlogged from the war years, and accommodated by a monetary policy which pegged interest rates at low levels until March 1951, aggregate demand remained robust until mid-1948, three full years after the war's end in August 1945.11 Even then, the decline in aggregate demand and the associated recession of November 1948 through October 1949 were modest and brief.
Balke, N. S., and R. J. Gordon. "The Estimation of Prewar Gross National Product: Methodology and New Evidence." Journal of Political Economy, February 1989, 38-92.
Bing, A. M. Wartime Strikes and Their Adjustment. New York: E. P. Dutton, 1921.
Board of Governors of the Federal Reserve System. Supplement to Banking and Monetary Statistics, Washington: Government Printing Office, 1966.
Brown, E. Cary. "Federal Fiscal Policy in the Postwar Period," in Postwar Economic Trends ill the United States, edited by Ralph E. Freeman. New York: Harper, 1960, 139-88.
Chow, G. C. "Tests of Equality Between Sets of Coefficients in Two Linear Regressions." Econometrica, July 1960, 591-605.
Douglas, P. H. Real Wages in tile United States 1890-1926. New York: Houghton Mifflin, 1930.
Economic Report of the President. Washington: Government Printing Office, 1979.
Freeman, Ralph E. "Postwar Monetary Policy," in Postwar Economic Trends in tile United States, edited by Ralph E. Freeman. New York: Harper, 1960, 51-90.
Friedman, M., and A. Schwartz. A Monetary History of the United States 1867-1960. Princeton: Princeton University Press, 1963.
Gordon, R. A. Economic Instability and Growth: The American Record. New York: Harper & Row, 1974.
Hickman, B. G. Growth and Stability of the Postwar Economy. Washington, D.C.: The Brookings Institution, 1960.
Hicks, J. R. A Contribution to the Theory of tile Trade Cycle. Oxford: Clarendon Press, 1950.
Hughes, J. The Government Habit. New York: Basic Books, 1977.
Klein, B. "Our New Monetary Standard: The Measurement and Effects of Price Uncertainty, 1880-1973." Economic Inquiry, December 1975, 461-84.
Lebergott, S. Manpower in Economic Growth. New York: McGraw-Hill, 1964.
Mock, J. R., and E. Thurber. Report on Demobilization. Norman: Oklahoma University Press, 1944.
Moore, G. H., ed. Business Cycle Indicators, Vol. I and II. Princeton: Princeton University Press, 1961.
Pilgrim, J. D. "The Upper Turning Point of 1920: A Reappraisal." Explorations in Economic History, Spring 1974, 271-98.
Romer, C. D. "The Prewar Business Cycle Reconsidered: New Estimates of Gross National Product, 1869-1908." Journal of Political Economy, February 1989, 1-37.
--. "World War I and the Postwar Depression: A Reinterpretation Based on Alternative Estimates of GNP." Journal of Monetary Economics, July 1988, 91-115.
Roose, K. D. "The Production Ceiling and the Turning Point of 1920." American Economic Review, June 1958, 348-56.
Soule, G. Prosperity Decode. New York: Holt Rinehart, 1947.
U.S. Council of Economic Advisers. Economic Report of tile President. Washington: Government Printing Office, 1991.
U.S. Department of Commerce. Historical Statistics of the United States. Washington: Government Printing Office, 1960.
--. Longterm Economic Growth, 1860-1970. Washington: Government Printing Office, 1973.
--. The National Income and Product Accounts of the United States, 1929-82. Washington: Government Printing Office, 1986.
U.S. Department of Labor. Monthly Labor Review, July 1924.
--. Monthly Labor Review, July 1929.
Wilson, T. Fluctuations in Income & Employment, 3rd ed. London: Pitman, 1948, chapters 11 and 12.
(1.) Only the Department of Commerce estimates are available for the Great Depression years. The Balke and Gordon  estimates and the Romer  estimates end with 1928.
(2.) See Moore  for business cycle reference dates [vol. I, 670], for the industrial production series [vol. II, 118], and for the factor employment series [vol. II, 129]. The source for monthly wholesale prices is U.S. Department of Labor [1924, 88-89].
(3.) For the purposes of this paper, the M1 money supply includes currency held by the public plus demand deposits in commercial banks. The M2 money supply is the M1 money supply plus time deposits in commercial banks. The monetary base is the sum of currency held by the public, bank vault cash, and commercial bank deposits at Federal Reserve banks.
(4.) The real wage data, taken from Douglas [1930, 391], end with 1926. The Lebergott [1964, 512] civilian labor force data begin with 1900, yielding 1901 as the first observation for percentage change.
(5.) One reason why the civilian labor force growth rate was larger for 1920 (4.1 percent) than for 1919 (1.6 percent), even though the decline in armed services personnel was somewhat greater for 1919 than for 1920, is that the 1919 influx of veterans was partly offset by an exodus of temporary civilian war workers--chiefly women, but also male teenagers and older men. Lags between demobilization and entry into the civilian labor force is another reason. The 1920 rise in immigration is still another reason.
(6.) The Bureau of Labor Statistics weighted wholesale price indexes including and excluding farm products both peak in May 1920, the increases over April being very slight. Bradstreet's unweighted index of wholesale prices of ninty-six farm and nmffarm commodities peaks in January 1920 (Moore [1961, 114, Vol. I]). The Mack unweighted index of fifteen industrial raw material spot market prices peaks in April 1920 (Moore [1961, 114, Vol. I ]).
(7.) Consistent with the idea of a supply side disturbance of some type being involved in the 1920-21 recession, the civilian unemployment rate rose by more during 1920-21 (5.2 percent to 11.7 percent ) than during the earlier 1907-8 (2.8 percent to 8.0 percent) and 1913-14 (4.3 percent to 7.9 percent) recessions even though the percentage decline in real product was about the same size or smaller during the 1920-21 recession, at least using the Balke and Gordon  and Romer  estimates. Using the Department of Commerce data, the percentage decline in real GNP was significantly larger during 1920-21 than during 191314, so that instance would say nothing. The unemployment rate could have risen by more in 1920-21 because aggregate product demand declined by more.
(8.) The source for wholesale prices excluding farm products and food, annually and monthly, is Moore [1961, vol. II, 145].
(9.) Most of the strikes were of short duration. For 1919, approximately 64 percent were of thirty-one days or less, and 90 percent were of ninty-one days or less (Bing [1921, 295]).
(10.) The survey results are reported in the Philadelphia Bulletin.
(11.) For discussions of the behavior of aggregate demand during the World War II aftermath, see Friedman and Schwartz [1963, 546-85], Hickman [1960, 34-78], Brown , and Freeman.
J.R. VERNON, Professor, University of Florida. The author wishes to thank three anonymous referees for helpful comments.
|Printer friendly Cite/link Email Feedback|
|Date:||Jul 1, 1991|
|Previous Article:||Customer discrimination and affirmative action.|
|Next Article:||Transaction cost, two-part tariffs, and collusion.|