Tuesday, January 24, 2012

US Unemployment in the 1890s

With all the talk of the recession of 1920–1921 at the moment (see here and here), there is another issue: the double dip recession of the 1890s.

Various Austrians are asserting that 1920–1921 proves that austerity can “quickly” end a recession. I have debunked that nonsense here, and the dishonest (or at least misleading) reference to a depression of 1920–1921, when there was no such thing, just a mild or moderate recession (depending on whether you use the revised data of (1) Romer or (2) Balke and Gordon).

Moreover, there was quite clearly a mild or moderate recession in the 1890s that completely contradicts the Austrians’ belief that austerity leads to rapid prosperity and high employment.

I. The GNP Data
According to the figures of Balke and Gordon, 1890s America suffered a double dip recession, with contractions in real GNP from 1893–1894 and 1896.

Balke and Gordon’s estimates for real GNP are here (the GNP growth rates are my own calculations):
Year | GNP* | Growth Rate
1890 | $183.9 | 1.43%
1891 | $189.9 | 3.26%
1892 | $198.8 | 4.68%
1893 | $198.7 | -0.05%
1894 | $192.9 | -2.91%

1895 | $215.5 | 11.7%
1896 | $210.6 | -2.27
1897 | $227.8 | 8.16%
1898 | $233.2 | 2.37%
1899 | $260.3 | 11.6%
1900 | $265.4 | 1.95%
* Billions of 1982 dollars
(Balke and Gordon 1989: 84).
As we can see, according to these figures, the US had a moderate recession from 1893–1894 in which GNP fell by 2.96%, with a recovery in 1895, but a further serious recession in 1896 with real GNP falling by 2.27%.

II. Unemployment
What were the effects of these output shocks on employment? There are three estimates that have been done:
(1) Lebergott’s estimates of the unemployment rate.
(2) Romer (1986: 31):
(3) Vernon (1994: 710).
Here are Lebergott’s estimates of the unemployment rate:
Year | Unemployment Rate
1890 | 4.0%
1891 | 5.4%
1892 | 3.0%
1893 | 11.7%
1894 | 18.4%
1895 | 13.7%
1896 | 14.5%
1897 | 14.5%
1898 | 12.4%
1899 | 6.5%
1900 | 5.0%
By these figures, the unemployment rates were a disaster in the 1890s, but Lebergott’s figures are challenged by Romer (1986).

The revised figures in Romer are as follows:
Year | Unemployment Rate
1892 | 3.72%
1893 | 8.09%
1894 | 12.33%
1895 | 11.11%
1896 | 11.965
1897 | 12.43%
1898 | 11.62%
1899 | 8.66%
1900 | 5.00%
(Romer 1986: 31).
Even using Romer’s figures, the US economy did not return to high employment for nearly a decade after 1893.

Finally, here are Vernon’s (1994) figures:
Year | Unemployment Rate
1890 | 3.97%
1891 | 4.34%
1892 | 4.33%
1893 | 5.51%
1894 | 7.73%
1895 | 6.46%
1896 | 8.19%
1897 | 7.54%
1898 | 8.01%
1899 | 6.20%
(Vernon 1994: 710).
They are lower than Romer’s, but still in the high single digits.

So it does not matter what figures you use: the double dip recession of the 1890s led to high unemployment that persisted to the end of the decade. There was a period of protracted unemployment in the 1890s comparable to the aftermath of the Great Depression (in the years from 1933–1939).

III. Conclusions
An important point is that 1890s America had no central bank, government spending was a very small percentage of GDP (it fluctuated between 2.55% and 3.62% in the 1890s), and governments tended to pursue austerity in times of recession. In fact, US federal government spending fell from 1893 to 1896 and fell from $465.1 million in 1893 to $443.1 million by 1896, which was obviously contractionary fiscal policy. Yet the culmination of the fiscal contraction in 1896 saw the economy in recession again.

Above all – and I wish to emphasise this – the fiscal contraction from 1893-1896 is correlated with rising unemployment in both the unemployment estimates of Romer and Vernon. Even by Vernon’s figures unemployment remained at nearly 8% until 1898. In Lebergott’s original estimates, unemployment soared from 1892-1894, went down in 1895, but then surged again in 1896 and stayed at 14.5% in 1897. No estimates of unemployment give any support to the view that austerity returns a shocked economy to high employment quickly. Curiously, a quick look at the data on
US federal government spending shows that spending rose from 1897 to 1899, and that this is also correlated with falling unemployment in the estimates from 1897 to 1899.

As an aside, I note how utterly absurd it is for Austrians to invoke 1920–1921 as an (alleged) vindication of their theories, when in that period America had a central bank. By any definition, 1920–1921 was even less of a laissez faire system than 1890s America, so it should be less relevant than the 1890s.

If 1920–1921 can be invoked as some kind of “proof” that austerity works, then, with even greater reason, the 1890s should show the “proof” of austerity too. But it does no such thing: although there was some high real GNP growth after the double dip in 1896, this was not sufficiently high to bring unemployment down.

Why was this? After all, real GNP growth rates of 8.16% (in 1897) and 11.6% (in 1899) seem very high by the contemporary averages of the mature US economy.

But there is a crucial issue: the US was a newly industrialising economy in the late 19th century, and in this respect was very much like China in the last three decades. With a large reserve of urban labour, coming from the countryside and from overseas in the case of the US in the late 1800s, an industrialising economy requires very high growth rates to maintain employment levels. In the case of China, a GDP growth rate of less than 7–8% leads to serious unemployment:
“‘China needs a growth rate of at least 7 per cent to avoid massive unemployment’ (www.economist.com, 10 November 2008). ‘The original estimated for China’s minimum rate of growth, which was made in the mid-1990s, was 7 per cent’ (The Economist, 15 November 2008, p. 88).

More recently somewhat higher figures for minimum GDP growth have been mentioned. ‘Most economists estimate that 8 per cent growth is needed to prevent urban unemployment from rising, which could trigger demonstrations and undermine the country’s social stability’ (www.iht.com, 20 October 2008; IHT, 21 October 2008, IHT, 21 October 2008, p. 11).

‘The government is expected to supply a fiscal stimulus to keep growth above 8 per cent’ (The Economist, 11 October 2008, p. 110). ‘China's own leaders believe they need growth of at least 8 per cent a year to avoid painful unemployment’ (The Economist, 15 November 2008, p. 14).” (Jeffries 2011: 10).
In other words, a growth rate of less than 7% in China today is the functional equivalent of a recession for workers in terms of its effects on unemployment.

I suspect a similar phenomenon was going on in 19th century America: just because there were positive growth rates (even what seem like high ones in 1897 and 1899), it does not mean that unemployment was always falling or that the economy was booming.

A research question I would propose is: what level of real GNP growth was necessary in 1890s America to mop up idle labour and reduce high unemployment? If there was a certain level of positive GNP growth required to prevent falling unemployment, a moderate recession (in technical terms) with a contraction of 2.96% in GNP may well have been a disaster for employment levels. In fact, it is possible that positive growth rates of 1%–4% may have been insufficient to maintain employment. All in all, this suggests to me that America’s actual GNP was well below its potential GNP in these years. This was not a healthy economy: it was an economy operating at well below its potential and no “proof” of the success of austerity at all. Rather, the 19th century, laissez faire policies of the US government were a disaster, above all in terms of unemployment.

It is no surprise to me that you do not see the Austrians appealing to the 1890s as an example of the wonders of the free market allegedly ending the aftermath of a recession, because on the metric of unemployment alone the 1890s completely contradict their absurd fantasies.

Appendix 1: Romer’s Figures for GNP in the 1890s

Romer’s estimates for real GNP are here (the GNP growth rates are my own calculations):
Year | GNP* | Growth Rate
1890 | $182.964 | 4.53%
1891 | $191.757 | 4.80%
1892 | $204.279 | 6.53%
1893 | $202.616 | -0.81%
1894 | $200.819 | -0.88%
1895 | $215.668 | 7.39%
1896 | $221.438 | 2.67%
1897 | $233.655 | 5.51%
1898 | $241.459 | 3.33%
1899 | $254.728 | 5.49%
1900 | $264.540 | 3.85%
* Billions of 1982 dollars
(Romer 1989: 22).
Romer’s figures show no contraction in 1896, and only a mild contraction in 1893–1894. Yet we know by all estimates unemployment soared in these years. What is going on? Romer’s estimates might be flawed. More likely, I think this supports the view that America in the late 19th century was very much like China today: just because growth rates were positive does not necessarily mean the economy was healthy, or that it was growing at its potential capacity.

Appendix 2: Were Movements in the Labour Force Pro-cyclical or Countercyclical in the 19th century?

There is the question whether movements in the labour force – especially involving women – were pro-cyclical or countercyclical in the 19th century. If it was countercyclical, this adds to unemployment, as women, young adults, and perhaps even children go out and look for employment when their husband/fathers/breadwinners lose employment (for literature, see James and Thomas 2007; Weir 1986, 1992). This is relevant for the method and accuracy of unemployment estimates in the 1890s.

BIBLIOGRAPHY

Balke, N. S., and R. J. Gordon, 1989. “The Estimation of Prewar Gross National Product: Methodology and New Evidence,” Journal of Political Economy 97.1: 38–92.

James, J. A. and M. Thomas, 2007. “Romer Revisited: Long-Term Changes in the Cyclical Sensitivity of Unemployment,” Cliometrica 1.1: 19–44.

Jeffries, I. 2011. Political Developments in Contemporary China: A Guide, Routledge, Oxon, England and New York.

Lebergott, S. 1964. Manpower in Economic Growth: The American Record since 1800, McGraw-Hill, New York.

Lebergott, S. 1964. Men Without Work: The Economics of Unemployment, Prentice-Hall, Englewood Cliffs, N.J.

Lebergott, S. 1986. “Discussion,” Journal of Economic History 46: 367-371.

Romer, C. D. 1986. “Spurious Volatility in Historical Unemployment Data,” Journal of Political Economy 94: 1–37.

Vernon, J. R. 1994. “Unemployment Rates in Post-Bellum America: 1869–1899,” Journal of Macroeconomics 16: 701–714.

Weir, D. R. 1986. “The Reliability of Historical Macroeconomic Data for Comparing Cyclical Stability,” The Journal of Economic History 46.2: 353–365.

Weir, D. R. 1992. “A Century of U.S. Unemployment, 1890–1990: Revised Estimates and Evidence for Stabilization,” Research in Economic History 14: 301–346.

11 comments:

  1. See my response on kuehns page, and mine which is awaiting moderation on murphys. And slowly we arrive at vernons unemployment figures, which show a much more modest unemployment rate. Funny how balkes GNP contractions are troublesome for the 1890s, but nothing of consequence for 20to 21. I have yet to read vernons paper, and am unsure of how he derives his figures, but my point in our debate, which you utterly refuse to mention still stands.

    ReplyDelete
  2. "I have yet to read vernons paper, and am unsure of how he derives his figures, but my point in our debate, which you utterly refuse to mention still stands.

    That is rubbish.

    Even by Vernon's figures unemployment soared.

    I note how Austrians today declare that America is in a depression because the unemployment rate is 9%.

    By the same criterion, America was also in a depression in the later 1890s slightly less bad than today, even by Vernon's estimates.

    By Romer's its depression was much worse.

    And I already answered your criticisms on GNP: it is clear that the US was a newly industrialising economy in the late 19th century, and in this respect was very much like modern China, as I have said above.

    ReplyDelete
  3. "Various Austrians are asserting that 1920–1921 proves that austerity can “quickly” end a recession. I have debunked that nonsense here"

    LOL you have done no such thing. All you did in that pathetic post was

    1. Argue over the definitions of "short" and "long." You asserted that an 18 month recession should be defined as "long", and not "short", without even explicitly defining what exactly constitutes "short" and "long".

    2. Fail to recognize the fact that ABCT tells us that the 1920-1921 recession would have been even longer if the government used monetary and fiscal stimulus to stop the needed corrections. ABCT makes no claims to actual time periods of boom or bust.

    3. Argue over the definition of "severe." You asserted that the recession should not be defined as "severe", contrary to the Austrians who defined that period as a "depression." Whatever you want to call it, it was not as "severe" as the Keynesian controlled Great Depression.

    4. Assert that falling prices based on increased production are somehow anything other than completely consistent with Austrian economics.

    5. Assert the silly belief that because there were no widespread bank failures, it somehow constitutes a counter-point to Austrian economics, when the reality is that because fractional banks in 1920-1921 were not so overextended so as to lead to a domino effect of bank failures across the economy, it vindicates the Austrians who argue against FRB.

    6. Fail to recognize that the economy recovered in July 1921. Monetary loosening of the Fed post May-1921, which always has a delayed effect on (false) economic recoveries, usually around 6 months to a year, depending on the specific circumstances, means that by the time the economy was recovering, it wasn't because of monetary easing. Monetary easing could only have generated a (false) recovery farther into the future, but by that time, the economy was ALREADY recovered, without Fed stimulus. The monetary stimulus of the Fed after May-1921, which was by the time the economy was already recovered, began a decade of loose monetary policy, which of course fueled the next artificial boom, and once the Fed tightened up again to avoid consumer prices getting too high, the economy again collapsed, and the Great Depression followed.

    ReplyDelete
  4. Let's look at your idiocy:

    (1) "You asserted that an 18 month recession should be defined as "long", and not "short", without even explicitly defining what exactly constitutes "short" and "long""

    The usual lies, I see.

    1920-1921 was long relative to the post-1945 business cycle, as I said quite clearly in the original post.

    Averages for US business cycles
    1919-1945 (6 cycles) 18
    1945-2009 (11 cycles) 11

    http://www.nber.org/cycles/cyclesmain.html

    The average length of US recessions from 1945-2009 is 11 months, including the great recession (December 2007- June 2009).

    (2) "Fail to recognize the fact that ABCT tells us that the 1920-1921 recession would have been even longer if the government used monetary and fiscal stimulus to stop the needed corrections. "

    I see! So in 1945-2009 when governments used monetary and fiscal stimulus recessions were shorter, yet somehow in 1920-1921 this would not have worked. More idiocy on display.

    (3) "Whatever you want to call it, it was not as "severe" as the Keynesian controlled Great Depression."

    Hoover didn’t use an effective Keynesian stimulus to end the contraction of 1929-1933, and anyone who says so is ignorant. In fiscal year 1930, Hoover actually ran a budget surplus, not a deficit. Federal policy was contractionary in this fiscal year.

    In fiscal year 1933, total federal spending was cut in relation to fiscal year 1932. Hoover introduced the Revenue Act of 1932 (June 6) which increased taxes across the board and applied to fiscal year 1932 and subsequent years. These were highly contractionary measures, and these two policies are the very antithesis of Keynesianism.

    So all you’re left with is fiscal year 1931 and 1932: Hoover did indeed raise federal spending in these years (especially in 1932), but it was woefully inadequate. In no sense do these miserable increases compared to the scale of the GDP collapse contradict Keynesian economics. Once you factor in state and local austerity and surpluses total federal spending barely moved.

    (4) This is straw man nonsense.

    (5) "Assert the silly belief that because there were no widespread bank failures,"

    There were no mass bank failures. That is a fact, and significant. Unless you believe the bank collapses in 1929-1933 had no effect on the economy, it is highly relevant to why 1920-1921 was not such a serious contraction.

    (6) "Fail to recognize that the economy recovered in July 1921."

    Laughable nonsense. The economy recovered some months after monetary loosing began, and contined to recover under further rate cuts.

    In all other circumstances, Austrians would be screaming that this would have caused a "false" recovery and an ABCT. But for some reason - it didn't in this case. This is absurd special pleading.

    ReplyDelete
  5. Lord keynes, can you post my reply

    ReplyDelete
  6. No.
    If you want further comments published then criticise the arguments of the post above, as it stands. I will then reply if I think they make a contribution.

    ReplyDelete
  7. "In fact, it is possible that positive growth rates of 1%–4% may have been insufficient to maintain employment."

    Unemployment rate is not a question of growth. Correlation is not (always) causation, you know...

    Oh, and about the interest rate cuts (of "only" -0.5 percentage point) occurred in May 1921 ...

    http://www.bankofcanada.ca/monetary-policy-introduction/why-monetary-policy-matters/4-monetary-policy/

    "The Bank of Canada’s policy actions relating to the overnight interest rate have almost immediate effects on the exchange rate and interest rates, but current estimates suggest that it takes between 12 and 18 months for most of the effect on aggregate output to be observed. Most of the effect on inflation is not apparent for between 18 and 24 months (Duguay 1994). And even these estimates are subject to considerable variation. [...] In particular, these long time lags mean that central banks must be forward-looking in their policy decisions. [...] If, on 1 January 2005, the Bank of Canada observes an event in the world economy that is likely to reduce aggregate demand beginning in June of the same year, there is nothing the Bank can do in January to fully offset that shock. Even if it responded immediately and lowered its policy rate in early January, there simply would not be enough time for its policy to stimulate aggregate demand sufficiently to offset the effects of the shock by June."

    ReplyDelete
  8. "Unemployment rate is not a question of growth. Correlation is not (always) causation,"

    I see! So you can get negative growth and rising unemployment, I suppose? Pray tell me how that happens. Magic?

    "Most of the effect on inflation is not apparent for between 18 and 24 months "

    (1) business expectations are subjective: a rate cut and the expectation of further rate cuts can induce hiring and borrowing for investment, the beginnings of investment and business confidence. You don't measure these things by mere inflation rates or even output indices.

    (2) And the recovery from 1921 continued in 1922, reaching a boom in 1923, as unemployment fell from its peak of about 8.8% to 5% by late 1922 - all completely consistent with the view that rate cuts were an important cause of the expansion.

    ReplyDelete
  9. "So you can get negative growth and rising unemployment, I suppose? "

    I don't think "unemployment rate" in a hunter-gatherer economy would be ten- or hundred- or thousand fold the usual rate of today modern economy. Slow- or negative-growth is a consequence, not causation. The slow growth may be due to a decline in aggregate demand (because of indebtedness), for example. The major cause of unemployment is "rigidity". Especially when prices are falling, and wages are sticky.

    "all completely consistent with the view that rate cuts were an important cause of the expansion."

    Again, correlation is not causation. The resolving of 1921 depression is due to a sharp decline in nominal wage. The economy immediately recovered after the post war restructuration.

    ReplyDelete
  10. You assume the Walrasian myth of wage and price flexibility causing market clearing (including the labour market). This is a fantasy.

    Anyway, it wasn't a short recession: it was 18 months.

    The GNP contraction may only have been 3.47% - a moderate recession, not a depression.

    There were a number of other factors that explain the lack of severity:

    (1) there was no huge debt-fuelled asset bubble collapsing
    (2) no massive, excessive private debt
    (3) no financial crisis or mass banking
    collapse.
    (4) positive supply shocks that benefited certain sectors of the economy.

    I've already exposed the Austrian nonsense about 1920-1921 here:

    http://socialdemocracy21stcentury.blogspot.com/2010/10/us-recession-of-19201921-some.html

    ReplyDelete
  11. "There were a number of other factors that explain the lack of severity [...]"

    And you forgot this one :
    (5) massive post-war restructuration. And this [restructuration] is what we, austrians, called ... a malinvestment. That malinvestment requires a "shift" in relative demand.

    ReplyDelete