There is a common belief among keynesians that a situation of a falling prices will result in a recession, that is, the business cycle is related to the fluctuations in prices. As we shall see, this claim is not supported by empirical evidence. Needless to say, a deflation resulting from a monetary contraction as a result of an increase in interest rate should not be confused with a deflation resulting from economic growth as a result of productivity gain. The first implies a stagnation or decline of economic growth, while the second implies an increase in economic growth where there are more goods to purchase. Regarding the “debtor-creditor injustice”, see Selgin (1988, ch. 9, pp. 106-107).
Selgin (1997) also relates the ‘Great Depression’ of 1873-1896 in UK : “Instead of inspiring large numbers of workers to go on strike, falling prices were inspiring them to go shopping”.
The United States and Great Britain both experienced a traditional immediate postwar boom, continuing the wartime inflation, in 1919 and 1920, followed by a severe corrective recession and deflation in 1921. The English deflation did not suffice to correct the overvaluation of the pound, since the United States, now the strongest country on gold, had deflated as well. The fact that sterling began to appreciate to the old par during 1924 misled the British into thinking that the pound would not be overvalued at $4.86; actually, the appreciation was the result of speculators betting on a nearly sure thing: the return to gold during 1925 of the pound at the old $4.86 par.
A crucial point: while prices and wage rates rose together in England during the wartime and postwar inflationary boom, they scarcely fell together. When commodity prices fell sharply in England in 1920 and 1921, wages fell much less, remaining high above prewar levels. This rise in real wage rates, bringing about high and chronic unemployment, reflected the severe downward wage rigidity in Britain after the war, caused by the spread of trade unionism and particularly by the massive new unemployment insurance program. 
The condition of the English economy, in particular the high rate of unemployment and depression of the export industries during the 1922–1924 recovery from the postwar recession, should have given the British pause. From 1851 to 1914, the unemployment rate in Great Britain had hovered consistently around 3 percent; during the boom of 1919–1920, it was 2.4 percent. Yet, during the postwar “recovery,” British unemployment ranged between 9 and 15 percent. It should have been clear that something was very wrong.
It is no accident that the high unemployment was concentrated in the British export industries. Compared to the prewar year of 1913, most of the domestic economy in Britain was in fairly good shape in 1924. Setting 1913 as equal to 100, real gross domestic product was 92 in 1924, consumer expenditure was 100, construction was 114, and gross fixed investment was a robust 132. But while real imports were 100 in 1924, real exports were in sickly shape, at only 72. Or, in monetary terms, British imports were 111 in 1924, whereas British exports were only 80. In contrast, world exports were 107 as compared to 1913.
The sickness of British exports may be seen in the fate of the traditional, major export industries during the 1920s. Compared to 1913, iron and steel exports in 1924 were 77.5; cotton textile exports were 65; coal exports were 80; and shipbuilding exports a disastrous 35. Consequently, Britain was now in debt to such strong countries as the United States, while a creditor to such financially weak countries as France, Russia, and Italy. 
It should be clear that the export industries suffered particularly from depression because of the impact of the overvalued pound; and that furthermore the depression took the form of permanently high unemployment even in the midst of a general recovery because wage rates were kept rigidly downward by trade unions, and especially by the massive system of unemployment insurance. 
Rohtbard (pp. 400-403) then goes on :
But it took little time for things to go very wrong. The crucial British export industries, chronically whipsawed between an overvalued pound and rigidly high wage rates kept up by strong, militant unions and widespread unemployment insurance, kept slumping during an era when worldwide trade and exports were prospering. Unemployment remained chronically high. The unemployment rate had hovered around 3 percent from 1851 to 1914. From 1921 through 1926 it had averaged 12 percent; and unemployment did little better after the return to gold. In April 1925, when Britain returned to gold, the unemployment rate stood at 10.9 percent. After the return, it fluctuated sharply, but always at historically very high levels. Thus, in the year after return, unemployment climbed above 12 percent, fell back to 9 percent, and jumped to over 14 percent during most of 1926. Unemployment fell back to 9 percent by the summer of 1927, but hovered around 10 to 11 percent for the next two years. In other words, unemployment in Britain, during the entire 1920s, lingered around severe recession levels. 
The unemployment was concentrated in the older, previously dominant, and heavily unionized industries in the north of England. The pattern of the slump in British exports may be seen by some comparative data. If 1924 is set equal to 100, world exports had risen to 132 by 1929, while Western European exports had similarly risen to 134. United States exports had also risen to 130. Yet, amid this worldwide prosperity, Great Britain lagged far behind, exports rising only to 109. On the other hand, British imports rose to 113 in the same period. After the 1929 crash until 1931, all exports fell considerably, world exports to 113, Western European to 107, and the United States, which had taken the brunt of the 1929 crash, to 91; and yet, while British imports rose slightly from 1929 to 1931 to 114, its exports drastically fell to 68. In this way, the overvalued pound combined with rigid downward wage rates to work their dire effects in both boom and recession. Overall, whereas, in 1931, Western European and world exports were considerably higher than in 1924, British exports were very sharply lower.
Within categories of British exports, there was a sharp and illuminating separation between two sets of industries: the old, unionized export staples in the north of England, and the newer, relatively nonunion, lower-wage industries in the south. These newer industries were able to flourish and provide plentiful employment because they were permitted to hire workers at a lower hourly wage than the industries of the north.  Some of these industries, such as public utilities, flourished because they were not dependent on exports. But even the exports from these new, relatively nonunionized industries did very well during this period. Thus from 1924 to 1928–29, the volume of automobile exports rose by 95 percent, exports of chemical and machinery manufactures rose by 24 percent, and of electrical goods by 23 percent. During the 1929–31 recession, exports of these new industries did relatively better than the old: machinery and electrical exports falling to 28 percent and 22 percent respectively below the 1924 level, while chemical exports fell only to 5 percent below and automobile exports remained comfortably in 1931 at fully 26 percent above 1924.
On the other hand, the older, staple export industries, the traditional mainstays of British prosperity, fared very badly in both these periods of boom and recession. The nonferrous metal industry rose only slightly by 14 percent by 1928–29 and then fell to 55 percent of 1924 in the next two years. In even worse shape were the once-mighty cotton and woolen textile industries, the bellwethers of the Industrial Revolution in England. From 1924 to 1929, cotton exports fell by 10 percent, and woolens by 20 percent, and then, in the two years to 1931, they plummeted phenomenally, cottons to 50 percent of 1924 and woolens to 46 percent. Remarkably, cotton and woolen exports were at this point their lowest in volume since the 1870s.
[…] The relative rigidity of wage costs in Britain may be seen by comparing their unit wage costs with the U.S., setting 1925 in each country equal to 100. In the United States, as prices fell about 10 percent in response to increased productivity and output, wage rates also declined, falling to 93 in 1928, and to 90 in 1929. Swedish wages were even more flexible in those years, enabling Sweden to surmount without export depression and return to gold at the prewar par. Swedish wage rates fell to 88 in 1928, 80 in 1929, and 70 in 1931. In Great Britain, on the other hand, wage rates remained stubbornly high, in the face of falling prices, being 97 in 1928, 95 the following year, and down to only 90 in 1931.  In contrast, wholesale prices in England fell by 8 percent in 1926 and 1927, and more sharply still thereafter.
The blindness of British officialdom to the downward rigidity of wage rates was quite remarkable. […] Finally, by the spring of 1929, Leith-Ross was forced to face reality, and conceded the point. At last, Leith-Ross admitted that the problem was rigidity of labor costs:
If our workmen were prepared to accept a reduction of 10 percent in their wages or increase their efficiency by 10 percent, a large proportion of our present unemployment could be overcome. But in fact organized labor is so attached to the maintenance of the present standard of wages and hours of labor that they would prefer that a million workers should remain in idleness and be maintained permanently out of the Employment Fund, than accept any sacrifice. The result is to throw on to the capital and managerial side of industry a far larger reorganization than would be necessary: and until labor is prepared to contribute in larger measure to the process of reconstruction, there will inevitably be unemployment. 
Leith-Ross might have added that the “preference” for unemployment was made not by the unemployed themselves but by the union leadership on their alleged behalf, a leadership which itself did not have to face the unemployment dole.
Now, Rothbard tells us that the 1873-1879 Long Depression was in fact a period of steadily growth. In his “A History of Money and Banking in the United States” (2002, pp. 154-155) he writes :
Orthodox economic historians have long complained about the “great depression” that is supposed to have struck the United States in the panic of 1873 and lasted for an unprecedented six years, until 1879. Much of this stagnation is supposed to have been caused by a monetary contraction leading to the resumption of specie payments in 1879. Yet what sort of “depression” is it which saw an extraordinarily large expansion of industry, of railroads, of physical output, of net national product, or real per capita income? As Friedman and Schwartz admit, the decade from 1869 to 1879 saw a 3-percent-per-annum increase in money national product, an outstanding real national product growth of 6.8 percent per year in this period, and a phenomenal rise of 4.5 percent per year in real product per capita. Even the alleged “monetary contraction” never took place, the money supply increasing by 2.7 percent per year in this period. From 1873 through 1878, before another spurt of monetary expansion, the total supply of bank money rose from $1.964 billion to $2.221 billion — a rise of 13.1 percent or 2.6 percent per year. In short, a modest but definite rise, and scarcely a contraction.
It should be clear, then, that the “great depression” of the 1870s is merely a myth — a myth brought about by misinterpretation of the fact that prices in general fell sharply during the entire period. Indeed they fell from the end of the Civil War until 1879. Friedman and Schwartz estimated that prices in general fell from 1869 to 1879 by 3.8 percent per annum. Unfortunately, most historians and economists are conditioned to believe that steadily and sharply falling prices must result in depression: hence their amazement at the obvious prosperity and economic growth during this era. For they have overlooked the fact that in the natural course of events, when government and the banking system do not increase the money supply very rapidly, free-market capitalism will result in an increase of production and economic growth so great as to swamp the increase of money supply. Prices will fall, and the consequences will be not depression or stagnation, but prosperity (since costs are falling, too) economic growth, and the spread of the increased living standard to all the consumers. 
Elsewhere, Rothbard (2002, pp. 159-161) writes :
The record of 1879–1896 was very similar to the first stage of the alleged great depression from 1873 to 1879. Once again, we had a phenomenal expansion of American industry, production, and real output per head. Real reproducible, tangible wealth per capita rose at the decadal peak in American history in the 1880s, at 3.8 percent per annum. Real net national product rose at the rate of 3.7 percent per year from 1879 to 1897, while per-capita net national product increased by 1.5 percent per year.
Once again, orthodox economic historians are bewildered, for there should have been a great depression, since prices fell at a rate of over 1 percent per year in this period. Just as in the previous period, the money supply grew, but not fast enough to overcome the great increases in productivity and the supply of products. The major difference in the two periods is that money supply rose more rapidly from 1879 to 1897, by 6 percent per year, compared with the 2.7 percent per year in the earlier era. As a result, prices fell by less, by over 1 percent per annum as contrasted to 3.8 percent. Total bank money, notes, and deposits rose from $2.45 billion to $6.06 billion in this period, a rise of 10.45 percent per annum — surely enough to satisfy all but the most ardent inflationists. 
For those who persist in associating a gold standard with deflation, it should be pointed out that price deflation in the gold standard 1879–1897 period was considerably less than price deflation from 1873 to 1879, when the United States was still on a fiat greenback standard.
[…] Before examining the critical decade of the 1890s, it is well to point out in some detail the excellent record of the first decade after the return to gold, 1879–1889. … Economic recordkeeping a century ago was not nearly as well developed as today, but a clear picture comes through nonetheless. The Encyclopedia of American Economic History calls the period under review “one of the most expansive in American history. Capital investment was high; . . . there was little unemployment; and the real costs of production declined rapidly.”
[…] The figures tell a remarkable story. Both consumer prices and nominal wages fell by about 30 percent during the last decade of greenbacks. But from 1879–1889, while prices kept falling, wages rose 23 percent. So real wages, after taking inflation — or the lack of it — into effect, soared.
Year – Unemployment Rate : Lebergott | Romer
1890 – 3.97% | 3.97%
1891 – 5.42% | 4.77%
1892 – 3.04% | 3.72%
1893 – 11.68% | 8.09%
1894 – 18.41% | 12.33%
1895 – 13.70% | 11.11%
1896 – 14.45% | 11.96%
1897 – 14.54% | 12.43%
1898 – 12.35% | 11.62%
1899 – 6.54% | 8.66%
1900 – 5.00% | 5.00%
1901 – 4.13% | 4.59%
1902 – 3.67% | 4.30%
However, Selgin, Lastrapes and Whites (2010) reported a lower figure. The unemployment rate averages 8% from 1893 to 1898.
As can be seen, unemployment rate was much lower during the 1880s than it was during the 1873-1879 and 1893-1898 periods. Yet, the drop in CPI was steeper during 1880s than it was during the 1893-1898 period. In addition, the drop in CPI was even steeper during the 1873-1879 depression than it was during the 1880s while the economic situation in the period from 1893 to 1898 was far worse than in the 1873-1879 depression. A closer look at the trend in the real income for the period running from 1867 to 1896 will make this clear. Refer to Chart 3 & 8 from Friedman and Schwartz (1963) below. The trend in real GNP (1982 dollars) from 1869 to 1904 – see Balke and Gordon (1988, Table 10) – shows a similar pattern : higher growth in 1873-1879 (+27.83%) than in 1893-1898 (+17.36%), while the latter period shows a higher growth than in 1882-1887 (+13.12%). In addition, Christina Romer (1986, Table 1) shows that the industrial production was keep growing between 1873 and 1879 and between 1880 and 1892, while stagnating between 1892 and 1897.
Overall, there was a huge deflation from 1873 to 1896 but unemployment rates did not keep pace with the secular decline in prices (Friedman & Schwartz, 1963, pp. 93-94). See also Ethel D. Hoover “Retail Prices after 1850” (1960, Table 1 & Table 8).
“Economic Growth With Price Deflation, 1873-1896” (1977) by Roger Elwood Shields gives further details. Chart IV-5 (p. 161) shows a comparison of four price indexes (1872-1896). The same pattern emerges :
Also, Milton Friedman and Anna J. Schwartz in their famous book “A Monetary History of the United States” (1963) reached the conclusion that the business cycle is not related at all to the fluctuations in prices (see Chart 8 and Chart 8 bis). Page 93, they wrote :
Was economic growth more rapid during the earlier period of declining prices or during the later period of rising prices? Unfortunately, the readily available figures do not yield a simple, clear-cut answer. Kuznets’ aggregate net national product in constant prices rises at the rate of 3.7 per cent per year from 1879 to 1897, and at the rate of 3.2 per cent from 1897 to 1914. This implies a rise in per capita net national product of 1.5 per cent a year for the earlier period, of 1.4 per cent for the later. However, the results of such a calculation are extraordinarily sensitive to the choice of dates: the use of 1880, 1896, and 1913, instead of 1879, 1897, and 1914, gives a rise in aggregate net national product of 2.6 per cent per year from 1880 to 1896 and of 4.4 per cent from 1896 to 1913. Inspection of the graph of net national product (see Chart 8, below) suggests little significant change in the rate of growth over the period as a whole, but rather a sharp retardation from something like 1892 to 1896 and then a sharp acceleration from 1896 to 1901, which just about made up for lost time. If this be right, generally declining or generally rising prices had little impact on the rate of growth, but the period of great monetary uncertainty in the early nineties produced sharp deviations from the longer-term trend. This evidence reinforces the tentative conclusion reached in the preceding chapter that the forces making for economic growth over the course of several business cycles are largely independent of the secular trend in prices.
A comparison with Chart 3 (Friedman & Schwartz) will make this clear :
Again, it is clear that the deflation in the period running from 1892 to 1898 was less severe than it was during the 1873-1879 depression. As noted earlier, the economic situation was more worrying during the 1892-1898 period. The link between real growth and prices is not clear. The authors (p. 42, footnote 40) have said that :
Kuznets’ figures for the United States give no clear indication whether output per capita grew more or less rapidly during the generally deflationary period before 1896 than during the generally inflationary period thereafter; the result obtained depends critically on the particular initial and terminal years used for comparison (see Chap. 3 below). According to available estimates of income per head in constant prices for the United Kingdom, the deflationary period was characterized by a definitely higher rate of growth than the later inflationary period.
Chapter 6, they have an interesting discussion (p. 242) on these three periods : 1873-92, 1892-1913, and 1920-1929. See their table 11.
Perhaps the most interesting feature of the comparison, as in our earlier comparison of the first two periods, is the difference between the behavior of money and real magnitudes during the periods of moderately stable growth. The rate of growth of real income, both total and per capita, was very similar; of prices, highly varied. Wholesale prices declined by 3½ per cent per year in the first period, rose by over 1½ per cent in the second, and declined by nearly 1 per cent per year in the third. Implicit prices declined by 2 per cent in the first period, rose by 2 per cent in the second, and were roughly unchanged in the third. Yet in all three stable-growth periods, total real income grew at a rate close to 3½ per cent per year and real income per capita at a rate between 1.3 and 2.0 per cent per year. These results reinforce our earlier conclusion that there seems to be no necessary relation between the direction of movement of prices over a period covering several business cycles and the corresponding secular rate of growth of real output. Apparently the steadiness of the price movement is far more important than its direction.
Historically, benign deflation has been the far more common type. Surveying the 20th-century experience of 17 countries, including the United States, Atkeson and Kehoe (2004, p. 99) find “many more periods of deflation with reasonable growth than with depression, and many more periods of depression with inflation than with deflation.” Indeed, they conclude “that the only episode in which there is evidence of a link between deflation and depression is the Great Depression (1929-1934).”
[…] There have in fact been other 20th-century instances in which deflation coincided with recession or depression in individual countries over shorter time intervals. In the U.S. this was certainly the case, for example, during the intervals 1919-1921, 1937-1938, 1948-1949 (Bordo and Filardo 2005, pp. 814-19), and, most recently, 2008-2009. It remains true, nonetheless, that taking both 19th and 20th-century experience into account, it is, as Bordo and Filardo (ibid., p. 834) observe, “abundantly clear that deflation need not be associated with recessions, depressions, and other unpleasant conditions.”
Although the classical gold standard made deflation far more common before the Fed‘s establishment than afterwards, episodes of “bad” deflation were actually less common under that regime than they were during the Fed‘s first decades (ibid., p. 823). Benign deflation was the rule: downward price level trends, like that of 1873-1896, mainly reflected strong growth in aggregate supply.
The authors (2010, pp. 3-4, 5, 14, 41-42) further discuss the gold standard :
As the first panel of Figure 1 shows, most of the decline in the dollar‘s purchasing power has taken place since 1970, when the gold standard no longer placed any limits on the Fed‘s powers of monetary control. – (pp. 3-4)
As Bernanke‘s remarks suggest, unpredictable changes in the price level have greater costs than predictable changes. Benjamin Klein (1975) observed that, although the standard deviation of the rate of inflation was only a third as large between 1956 and 1972 as it had been from 1880 to 1915, inflation had also become much more persistent. The price level had consequently become less rather than more predictable since the Fed‘s establishment. – (p. 5)
Whereas one might expect the Fed, in its role as output stabilizer, to tighten the money supply in the face of positive IS (spending) shocks and to expand it in response to positive shocks to money demand, the response functions we estimate indicate instead that the Fed has tended to expand the money stock in response to IS shocks, causing larger and more persistent deviations of output from its “natural” level than would have occurred in response to similar shocks during the pre-Fed period (Figure 7, left-hand-side panels). At the same time, the Fed was less effective than the classical gold standard had been in expanding the money supply in response to unpredictable reductions in money‘s velocity. – (p. 14)
A fiat standard can in principle replicate a gold standard‘s price-level stability without any such resource costs (Friedman 1953). In practice, however, fiat standards have not replicated gold‘s price-level stability (Kydland and Wynne 2002, p. 1). Nor, ironically, have they even lowered resource costs. The inflation rates of postwar fiat standards have by themselves imposed estimated deadweight costs greater than the reasonably estimated resource costs of a gold standard (White 1999, pp. 48-49). Meanwhile, the public has accumulated gold coins and bullion as inflation hedges, adding more gold to private reserves than central banks have sold from official reserves. The real price of gold is much higher today than it was under the classical gold standard, encouraging the expansion of gold mining (Figure 12). Thus the resource costs of gold extraction and storage for asset-holding purposes have risen since the world‘s departure from the gold standard. – (pp. 41-42)
Concerning economic growth, Rothbard (2002, p. 164) provides some figures about the productivity growth between 1870s and 1880s, showing that american economy was growing at one of the fastest rates in its history despite the deflation at that time.
By some measures the 1880s was the most productive decade in our history. In their A Monetary History of the United States, 1867–1960, Professors Friedman and Schwartz quote R.W. Goldsmith on the subject:
The highest decadal rate [of growth of real reproducible, tangible wealth per head from 1805 to 1950] for periods of about ten years was apparently reached in the eighties with approximately 3.8 percent.
Keynesians may however argue that the period running from 1945 to 1973, known as the Classical Keynesianism, was also extremely productive when these economies were centrally planned (see Paul Davidson’s Global Employment and Open Economy Macroeconomics cited in Johan Deprez’s Foundations of International Economics: Post-Keynesian Perspectives; Table 2.1). Unfortunately, the comparison does not hold. Economic growth is much easier when the economy has to recover from the damages caused by the war. As Landier & Thesmar pointed out in “Le grand méchant marché” (p. 110) :
When it comes to rebuild, import and implement existing technologies, there is no uncertainty about the way forward, the point of arrival is known. Investments to accomplish and their relative importance are visible and consensual …
Even if the periods in question are comparable, other important factors affecting economic growth are not necessarily constant across time periods, especially for two very distant periods (1870s and 1950s). Since the 1870s, many improvements have been made in technology, institution, with a surge in scientific achievement, and other factors contributing to economic growth. Hence, we should keep these variables constant. Or we would rather say that economic growth could be higher than what it would be otherwise. Empirically, it has been proved that economic freedom affects economic performance (through economic freedom, achievement in technological and scientific research) even though intelligence is more important. See Heiner Rindermann (2008a, pp. 136-137, figures 4 & 5; 2008b, p. 316, section 5.4, and figure 7; 2011, figures 4 & 6; 2012, p. 110, and figures 1, 2 & 3).
As for the causes of the panics of 1873, 1884, 1893, and 1907, Rothbard attributes the crisis to the practice of the fractional reserve banking (2002, pp. 154, 160, 168, 240). However, George Selgin has a more convincing explanation in “The Theory of Free Banking: Money Supply under Competitive Note Issue” (1988, ch. 1 & 9). The root causes stem from the incapacity of the banking system to satisfy the increase in the public demand for money due to the banking regulation at that time. Rothbard (2002) gives a complementary explanation for the economic disturbance during the 1890s (see pp. 168-169, 184-186).
Further reading :
America’s Greatest Industrial Transformation, by Jonathan M. Finegold Catalán