The 1937-38 crash was literally a depression within a depression1. The seasonally adjusted production index peaked 118 in May 19372. A year later it stood at 76, a drop of 36 per cent. From April 1937 to May 1938 manufacturing output fell by 38 per cent. The situation for the iron and steel industry was catastrophic with output collapsing by 67 percent. Factory employment dived by 25 per cent, factory payrolls by 36 per cent while aggregate unemployment peaked at 20 per cent. Such a rapid contraction in production was and is unprecedented in US History. The statistics for manufacturing, and the iron and steel industry in particular, are both striking and instructive if the monthly production figures are examined instead of annual aggregates, a fact that will become increasingly clear.
Even today there is no consensus on the cause of the contraction. I realise that for most people this historical event is just that — history, an incident that has no bearing on the present. But the vast majority of the population — including our politicians and so-called media — do not appreciate the fact that the insidious fiscal and monetary policies that gave us the current economic situation have their roots in the Great Depression. Until it is properly understood what brought about these economic disasters we will continue to be cursed by the same economic policies that brought us the present sorry economic state.
There are a number of explanations for the 1937-38 crash. For example, Alan Moran of the Institute of Public Affairs has adopted the view that “once debt creation was slowed in 1937, the economy again tanked”. (The Looming Disaster from Deficit Spending, October 2013). Dr Moran overlooked the fact that from 1932 to 1937 Commonwealth debt actually fell by about 3 per cent. Instead of tanking the Australian economy continued to expand and create more and more jobs. This is proof positive that correlation is not causation. On the other hand, if Dr Moran means by “debt creation” credit expansion then this is dealt with below.
In reality there are only three explanations of the crash that contain any real substance. The first has Milton Friedman and Anna Schwartz blaming the monetary contraction for the tragedy. (This could be Alan Moran’s “debt creation” solution to the riddle). A considerable number of people still adhere to this explanation. However, anti-monetarists, mainly Keynesians, point the finger at the drop in government spending, claiming that this fiscal contraction reduced aggregate demand and that this triggered the drop in production. A third group argues that the magnitude of the depression can only be explained by the huge wave of strikes that resulted in wage rates that savaged price margins and recklessly drove up the cost of production.
I believe the third explanation is the correct one. To make a case I shall deal with each one in some detail, starting with the Friedman-Schwartz account. Prices had been rising and the Roosevelt administration was worried that inflation was taking root. In response to this fear the fed decided to raise reserve requirements in three stages. In August 1936 they raised reserve requirements by 5 per cent. The next two increases were on 1 March and 1 May 1937, thereby doubling the reserves. (Although reserves were doubled the requirements for the banks were not uniform. Nevertheless, the overall result was the same.) This action reduced reserves from $3 billion to about $0.93 billion. Monetarists therefore deduced from this that the new requirement cut funding to business to the extent that it triggered a deflation which in turn precipitated the 1937 crash.
The monetarists assume far too much. In doing so they overlooked a great deal of contrary evidence. Classical economists observed that whenever the banks created a sudden deflation there would be a rapid and desperate rush by businessmen for cash resulting in interest rates for short term loans being driven rapidly up. Robert Mushet, a nineteenth century economist, witnessed rates for businessmen jump as high as 40 per cent to 60 per cent as a result of the 1825 economic crash caused by the Bank of England’s desperate effort to restore its exhausted reserves3. This is precisely what one would expect in these circumstances.
The stress is on the restoration of reserves. After a series of reckless monetary expansions followed by steep deflations the English banks once again embarked on another irresponsible monetary expansion in 1822 that created a wave of frenzied speculation and eventually generated an external and internal gold drain4 so severe that by late 1825 the banks’ reserves had fallen to a point where the banking system found itself on the edge of a financial precipice. The situation was so grave that it nearly brought about the fall of the Bank of England. The result was a swift and dramatic deflation that brought on a deep depression.
Now this was obviously not the case in 1936 or 1937. The fact remains that the system had been holding excess reserves since 1929. By 1940 member bank reserve balances had grown to $13,249 billion while excess reserves stood at $6,326 billion. This is why there was no frantic scramble by business that would have driven up short term rates. A monetarist could point to the fact that immediately the money supply peaked in March the rate on prime commercial paper rose by 33per cent in April. The commercial paper rate is important because it is the one that reflects the short term rate that business borrowed on. From February 1935 the rate had remained unchanged at 0.75 per cent. After rising a quarter of one per cent, it stayed there until February 1938, after which it began a slow decline.
Even when the rate rose to 1 per cent it was still barely positive. Moreover, it was much lower that the rates that prevailed in the booming 1920s and in the post-WWII economy. If monetary tightening had caused the crash then this rate should have risen significantly. In addition, Prime bankers’ acceptances stayed under 1 per cent, where they had been since May 1933. Ninety-day stock exchange time loans remained unchanged at to 1.25. These rates were absurdly low and for anyone to suggest that raising any of them by a fraction of 1 per cent would have generated the fastest drop in production and employment in US history is just plain ridiculous. Equally ridiculous is the thought that a 6.4 per cent monetary contraction, which is what occurred, could have the same effect.
The most damaging evidence against the monetary interpretation of the crisis is that the thing is out of sequence with the pattern of production. Money supply peaked in March 1937 and industrial production peaked in May at 118 where it stayed until falling to 114 in June, finally plunging to 80 by January 1938. Now I am second to none in my respect for the power of money to disrupt an economy but I just cannot see how industrial production could have responded so quickly to the change in the money supply, especially when nothing else did.
Moreover, when we examine industrial production in greater detail the case against the monetary explanation becomes overwhelming. Although industrial production peaked in May manufacturing peaked in April, only several weeks after the money supply peaked. This is a little too fast. The clincher is that iron and steel production peaked in January, two months before M1ceased expanding5.
Roosevelt’s devaluation of the dollar created a large inflow of gold. To prevent what they thought would be an inflationary surge the Treasury sterilised the inflow rather than allow it to add to the money supply. Critics of this policy argued that it contributed greatly to the tragedy. Two points here. First, it was a bit rich for these critics to attack the Treasury for not applying the gold-standard rule to the gold inflow when they were part of the chorus that demanded the demonetisation of gold. Second, it literally did not matter anyway because the system was holding excess reserves that were actually expanding. Monetising gold would simply have added to the excess. These critics also overlooked the fact that sterilising the gold could only have tightened the money market and hence raised interest rates if the banking system was in need of rebuilding reserves. In plain English, the sterilisation process had no effect on the supply of funds for business or the interest rate structure.
For the monetarist every significant deflation (monetary contraction) must be immediately followed by a contraction in output irrespective of the circumstances. This mechanistic view blinded them to the actual sequence of events. By sheer coincidence a similar monetary situation also occurred in Australia. When the first deflation ended in September 1931 it was immediately followed by a rapid reflation that came to an abrupt halt in March 1932. M1 then contracted, falling by 10.2 per cent by September 1933 when reflation once again set in6. The economy did not even flinch.
The Australian recovery continued uninterrupted with manufacturing leading the way7. By 1934 employment in manufacturing had reached its 1929 level and by 1938 was 25 per cent higher and still expanding. Chart 1 compares monetary changes in Australia with changes in the unemployment rate while chart 2 does the same for the US. It can be seen that the second Australian deflation was deeper than America’s 1937-38 deflation though shorter by four months. I just cannot imagine how a monetarist could explain this phenomenon.
The fiscal explanation for the Roosevelt depression fares no better even though chart 3 appears to confirm it by showing that changes in unemployment corresponded to changes in federal spending for the financial years 1928-29 to 1939-40. As always, the devil is in the details.
To get a clearer picture of the spending situation we need to look at it on a monthly basis. Chart 4 takes the monthly figures for the calendar years 1935 to 19388. Spending for year 1936 increased by 13.7 per cent over the previous year while for 1937 it fell by 3.2 per cent. As for 1938, spending scarcely rose at all. This is particularly interesting because all the economic indicators picked up in June of that year, despite the fact that federal spending had only increased by a meagre $147.5 million during the second half of 1937. It is scarcely likely that such a sum would ignite a rapid recovery from a severe contraction.
What is even more interesting than the June pick up in production is that anyone could suggest that a 3.2 per cent fall in spending could trigger a massive drop in production. If, however, we examine spending for the relative financial years we get a much grimmer picture. Although total spending (federal + state + local)9 continued to increase during this period, albeit slowly, chart 5 shows federal spending dropping by 12.3 per cent, despite the trend line. If we accept the premise that federal spending is the key variable then this chart certainly appears to confirm the Keynesian view.
Now the first thing to note is that the charts have two large spikes. These were due to the soldiers’ bonuses that were paid in June 1936 and June 193710. These massive payments clearly distorted the trend in government spending and explain why both charts produce different results. So which chart should be used? In my view, it is the calendar years that should count. Nevertheless, the issue has to be resolved in a way that definitively decides the matter.
Economics is neither physics nor chemistry, meaning that economists cannot carry out laboratory-like experiments. But they have the next best thing to replication: they have at their disposal a significant array of case studies upon which they can draw. These studies are in fact a record of historical events that allow us to compare the results of a current economic policy with the same or similar policies that were made in another decade or even era.
In short, the solution to the problem rests with applying a historical perspective. The 1920-21 depression fits the bill. From June 1920 to January 1922 wholesale prices fell by 44 per cent. The story was slightly different for retail prices. From June 1920 to December 1922 these fell by 22 per cent. It was particular grim for personal incomes which fell 15 per cent for the period 1920-21, national income dived by 19 per cent while the official unemployment rate rose to 11.7per cent. Manufacturing was hit the hardest with production contracting by 34 per cent from its peak in February 1920 to its trough in January 1921. If ever there was a dire need for a massive programme of deficit spending to save the economy, this was it.
As we can see from chart 6 government spending went from $18,514,880 in 1919 to $6,403,344 in 1920, a 65 per cent drop while for the same period a massive $13,370,638 deficit —17 per cent of GDP — was transformed into a $212,475 surplus. (The figures are in billions. Statistical Abstract of the United States 1942). Moreover, spending continued to fall until 1924 after which there was a slight increase. The 1920 crisis was signalled when manufacturing suddenly ceased expanding in February. Now what is really interesting is that manufacturing began its recovery in February 1921 while government spending was still contracting and surpluses were accumulating. (This is exactly what happened in Australia during the Great Depression).
According to the Keynesians this is just not possible. Of course it can be argued that this event is far from being clear cut because of the distortions carried over from the war, the existence of massive excess inventories, the drop in demand from Europe, the monetary contraction and so forth. But the important thing to note is that the recovery, along with the monetary expansion, followed the classical pattern. This was the last American recovery to do so11.
Fortunately for us Australia provides an absolutely clear cut case that no Keynesian can refute, no matter how much smoke he blows. Australia suffered a severe deflation with the official unemployment rate peaking at 30 per cent in June 1932. The government’s response to the disaster was to implement a policy of spending cuts and budget surpluses, the very policy Keynesians would condemn today as economically insane. The result was a recovery that put the Roosevelt administration and its spending policies to shame.
Chart 7 shows a Commonwealth (Fed to American readers) surplus for the financial year 1931-32 with unemployment peaking at 29 per cent. From there on unemployment continued to fall, with a slight reversal for the year 1938-9, until the country entered the war, despite the government running surpluses for the entire period. According to Keynesians, however, running surpluses during a deep depression will have devastating consequences. The Australian experience clearly exposes the Keynesian view as a complete fallacy.
Chart 8 is equally damning. It shows a steep fall in the unemployment rate for manufacturing workers12, even as Commonwealth and total spending continued to contract. According to Keynesians, this simply cannot be. Unfortunately, many member of our free-market commentariat jumped to the erroneous conclusion that the cuts were largely responsible for the dramatic fall in unemployment13.
The chart shows Commonwealth spending continuing to fall throughout the financial year 1933. It did not start rising again until the financial year 1934, by which time unemployment had fallen by about 33 per cent. Moreover, chart 8 also reveals that total spending peaked in the financial year 1931 and did not start rising again until the financial year 1935. (Even though Commonwealth and total spending eventually rose again, GDP rose faster).
Compare the situation with that in America. The unemployment rate officially peaked in 1933 at 25 per cent. By 1935 the rate stood at 20 per cent, notwithstanding the fact that Roosevelt had enthusiastically adopted the new economics of deficit finance. Yet, according to Keynesians an Australian approach to the depression would have made a disastrous situation worse by savaging government net income-creating expenditures.
Although the cuts may have contributed to Australia’s recovery (I believe they did) the real explanation lies with ‘changes’ to real wage rates. Chart 9 compares the real wage of factory workers (the money wage adjusted for changes in the price level) in Australia with that in America. Here lies the key to Australia’s recovery and Roosevelt’s failure. Whereas there was a continuous increase in real wage rates for American factory workers real wage rates for their Australian counterparts largely remained flat throughout that period, standing at only 102 in 194015.
Now chart 9 could suggest to most readers that there is no connection between the real wage rate and the demand for labour16. The error here is to overlook the fact that the real wage is not what determines the demand for labour. When hiring labour employers do not take account of the wage rate as divided by the price level. Their concern is what they have to pay in money terms with respect to what they calculate they will receive in nominal revenue from the services of the labour they hire. Therefore, if the price of labour services is raised above the value of what labour produces then unemployment will follow. In short, it is the ratio of the money wage rate to the value of the output that matters to the employer. We may call this ratio the real factory wage. Chart 10 brings this fact home with startling clarity.
The chart clearly shows that the inverse correlation between the real factory wage and the value of output is virtually perfect. This is basic economics. Raise the cost of labour above the value of its services and unemployment will emerge. The more labour costs are raised the more unemployment we get. Chart 11 is just as telling as chart 10 and explains why Australia’s unemployment record during the Great Depression was greatly superior to America’s.
Australia, like America, also suffered a massive deflation. From March 1929 to September 1931 M1 fell by 27.2 per cent with demand deposits dropping by 33 per cent. As we see from the chart, factory employment dived by 25 per cent and the value of factory output fell by 35 per cent. Note that the real factory wage peaked at 130 in the middle of 1931 and then began a slight decline. It was then that the contraction in factory employment came to a halt. Note also that at the same point there was a distinct slowdown in the rate at which the value of factory output was shrinking. During this period the money factory wage continued to fall until about June 1934 when it started to rise. The critical point is the middle of 1932 when the value of factory output started to rapidly increase and continued to do so throughout the remainder of the 1930s. (Not included in the chart was the dramatic 35 per cent drop in the prices of raw material inputs for manufactures).
Charts 10 and 11 provide clear proof that the America’s high and persistent unemployment was the result of excessive wage rates revealed by the real factory wage. This thesis is validated by the Australian experience. By excessive wage rates it is meant those in excess of the marginal value of the employee’s product and not his average productivity. These are entirely two different things. The principle point is very simple but rarely explained. It is blindingly obvious that if the value of a worker’s marginal product (the value of an additional unit of output) is x dollars but, let us say, unions insist on x+y dollars then unemployment must eventually emerge. It also follows that the greater the gap between full employment wage rates and union-imposed rates the higher will be the level of unemployment.
In America money wages in manufacturing had been rising slowly through 1935 and up to October 1936. Then after the November there was an explosion of union activity. To extract higher wage rates from employers the union leadership organised a wave of strikes, violent confrontations, and mass sit-downs. This was combined with a union recruiting drive that virtually unionised the whole of manufacturing. The result was wage rate increases that gave America the depression within a depression.
The average manufacturing wage in 1936 was $22.60, average weekly hours of work were 39.1 and the hourly rate was $0.564. Thanks to the union-led wage push the hourly rate had increased in 1937 by 12.4 per cent, raising the weekly wage to $24.95. Employers immediately responded by reducing the average working week to 38.6 hours. (Reducing working hours in these circumstances creates a form of hidden unemployment17). The situation was even worse in 1938 with the working week being further cut to 35.5 hours, dragging the weekly wage down to $22.70 and raising the level of hidden unemployment.
When we look at some individual industries we can see just how destructive the union wage push really was. The steel industry was a huge target (we should say victim) of the union leadership. Its average wage 1936 was $27.42 and the hourly rate was $0.671 cents while weekly hours of work were 40.9. The following year the unions had rammed the hourly rate up by 22 per cent. The weekly wage jumped to $31.64 and working hours fell to 38.7. (One should note that the fall in working hours means that output was contracting). The full impact on jobs and wages was revealed in 1938 when the average manufacturing wage had fallen to $23.92 and working hours to 28.718.
To get some idea of how a surge in excessive wages rate (still bearing in mind that by excessive we mean in excess of the value of the marginal product) let us imagine a manufacturer with a slim net return of 6 per cent and total operating costs of $100,000,000 with $30,000,000 being wages. A 22 per cent jump in the wage rate would wipe out his price margin and immediately put him in the red. Although this is a hypothetical example it is basically what happened to the steel industry. In fact, once we include the benefits of union-enforced work rules, pay roll taxes and other Roosevelt imposed burdens the total increase in labour costs must have exceeded 25 per cent. The same holds for other industries.
Let us look at several more industries. The hourly rate in cast-iron production was pushed up by 11.2 per cent: the weekly wage rose from $18.99 in 1936 to $21.17 in 1937, weekly hours fell from 38.2 to 37.8. In 1938 the hourly rate was now 15 per cent over the 1936 rate, but the weekly wage had fallen to $19.15 and the working week to 32.8 hours. It was the same story for steam systems and fittings. In 1936 the average weekly wage was $24.25 and working hours averaged 42.2. The following year saw the hourly rate driven up by 13.2 per cent and the weekly wage rise to $25.02 while hours of work dropped to 40. The hourly rate in 1938 was raised again so that it stood at 19.4 per cent over the 1936 rate. The average wage now fell to $23.15 while working hours dived to 33.1, creating even more hidden unemployment.
The average wage in 1936 in the foundry and machine-shop products industry was $25.56, weekly hours were 42.4 and the hourly rate was $0.601 cents. In 1937 the unions lifted the hourly rate by 13 per cent. They lifted it again in 1938 so that it was now 18.3 per cent above the 1936 rate. The weekly wage was now $24.94 but the working had been cut to 35 hours. Weekly wages in the machine tools industry were $28.40 in 1936, the hourly rate was $0.636 cents and the working week averaged 44.6 hours. By 1938 the unions had succeeded in raising the hourly rate by 20 per cent. Employers responded by lowering the working week to 36.3 hours, bringing the weekly wage down to $26.61. We find the same thing with aluminum (sic) manufactures. The industry’s average wage for 1936 was $23.38 and the working week was 41.3 hours. In 1938 the hourly rate was up by 20 per cent, working hours were down to 36.3 but the average weekly wage was now $24.07. (The figures are from the Statistical Abstract of the United States 1939, p. 329).
The 1937-38 depression was not caused by demand deficiency nor was it caused by a money contraction or a reduction in federal spending. The sole cause of the disaster was a brutal wage push by a militant union leadership. Huge percentage increases in wage rates savaged company price margins causing a massive contraction in production. It was this that triggered the monetary contraction that monetarists mistakenly think caused the crash. It is true that some historians point to Roosevelt’s destructive undistributed profits tax, the pay roll tax and other imposts as at least contributing factors. Damaging as these were there is absolutely no way they could have caused the crash. For instance, the undistributed profits tax was a direct tax on capital accumulation, a fact that an economic illiterate like Roosevelt was unable to grasp. This tax combined with the other factors would simply have kept the economy depressed.
Charts 10 and 11 conclusively show the vital relationship between the real factory wage and the value of manufacturing output. (Australia’s real factory wage fell by nearly 43 per cent from its peak of 130 in 1931 to 75 in 1939). Now the Australian experience demonstrates with indisputable clarity that no matter what happens to government spending unemployment can continue to fall so long as the ratio of the wage rate to the value of the output is allowed to decline. Although it is not immediately apparent the ratio approach confirms the marginal productivity theory of wages. If the theory was false then there would be no inverse correlation between the real factory wage and the value of manufacturing output that charts 10 and 11 reveal.
The standard economic theory of wages stands vindicated. Therefore, the solution for the problem of persistent wide-spread unemployment is clear. Labour costs must be brought into line with the marginal value of the product, which is what the Australian experience amply demonstrates. There are really only two ways in which this can be done19: increase productivity to the extent where it continues to reduce the ratio until full employment is restored or use inflation to cut real wage rates20. The latter is achieved by using inflation to raise the money price of the product relative to the wage rate. This is the Keynesian solution.
Oddly enough, the vast majority of Keynesians are genuinely oblivious to the fact that their policy proposal actually confirms the classical view that they believe Keynes refuted. They are also equally oblivious to the fact that the classical economists were aware of this deception and its dangers20. These economists fully understood that purchasing power springs from production (Say’s law) and this is why they rightly attacked as extremely dangerous the economic fallacy that consumer spending drives an economy. In 1937 three economists pointed out what should have been obvious to the profession as a whole:
The larger number of payments is not from consumers to producers, but is made between producers and producers, and tends to cancel out in any computation of net income of net product value. “In fact, income produced or net product is roughly only about one-third of gross income.” [Italics added]. What is cost for one producer is in part income for some other producer, but part of that income the latter has to pay out in costs to other producers in another stage of the productive process (for intermediate products, raw materials, supplies, etc.), and so on21. All that is necessary in order that equilibrium be maintained is that consumers’ incomes equal the cost of producing consumers’ goods; the total of producers’ payments necessarily exceeds that of consumers’ incomes. (C. A. Phillips, T. F. McManus and R. W. Nelson, Banking and the Business Cycle, Macmillan and Company 1937, p. 71).
John Stuart Mill used what he called the Fourth Proposition of Capital to express this fact in another way when presenting the classical case against what we now call Keynesian thinking:
Demand for commodities is not demand for labour. The demand for commodities determines in what particular branch of production the labour and capital shall be employed; it determines the direction of the labour; but not the more or less of the labour itself, or of the maintenance or payment of the labour22. (Principles of Political Economy, Vol II, p. 78, University of Toronto Press 1965.)
Once we grasp the fundamental truth that the classical economists understood then we can see why Roosevelt’s New Deal economics were an utter failure and why by 1938 Australian unemployment had fallen from its peak of 30 per cent in 1932 to 8.8 per cent while American unemployment stood at 20 per cent against its peak of 25 per cent in 1933. The comparison between factory employment in both countries is even more striking.
In 1938 US manufacturing employment was 14 per cent below its 1928 level while in Australia it was 26 per cent higher than the 1928 rate and still growing. Moreover, during the depression hours of work in Australian manufacturing never fell below 44. In 1938 the average working week for US manufacturing was 35.5 hours against 44.82 hours in Australia. (This figure covers all industrial groups except shipbuilding). Furthermore, weekly hours of work in American manufacturing clearly indicate that the real unemployment rate certainly exceeded the official figure.
The extent to which Australia took (albeit unintentionally) a near-classical approach to the depression explains why she did so much better than Roosevelt’s America.
Note: Also see Australia and the Great Depression. In a forthcoming article on the Great Depression in America I shall return to the 1937-38 crash.
* * * * *
1According to President Roosevelt and some of his supporters the second depression was caused by monopoly pricing, a thoroughly embarrassing explanation that his admirers would rather forget.
2Federal Reserve Bulletin, September 1939, p. 804
3Robert Mushet, An Attempt to Explain from the Facts the Effects of the Issues of the Bank of England Upon its Own Interests, Public Credit and the Country Banks, Baldwick, Craddock, and Joy, Paternoster Row, 1826, p. 179.
4Thomas Tooke dealt with this monetary episode in considerable detail in his Considerations on the State of the Currency (London: John Murray, Albemarle-Street, 1826)
5The pattern of decline is particularly interesting. Although industrial production fell by 36 per cent durables fell by 52 per cent, non-durables by 24 per cent and iron and steel by a staggering 67 per cent. Therefore, we find that not only is the pattern of the contraction out of sequence with the monetary explanation so is the sheer scale of the collapse.
6The monetary figures are from C, B, Schedvin’s Australia and the Great Depression, p. 386. The odd thing about this is that Australian economists and historians completely missed the second deflation.
7Australia’s recovery in manufacturing was led by “metals and machinery”. (Australia and the Great Depression, pp, 304- 5). The opposite case held for the US where manufacturing was consumption-led, causing severe imbalances in the production structure. By 1936 Australian manufacturing employed more people than before the depression and was still expanding capacity and creating even more jobs. The expansion was still proceeding when war was declared.
8Calendar year figures from the Statistical Abstract of the United States, 1937.
9The state and local spending figures came from www.usgovernmentdebt.us (now archived) which relies on official sources for its data.
10To fund the soldiers’ bonuses the government issued bonds worth about $2 billion. Most of these were issued in June 1936. The June expenditures also included $500 million of bonds that were deposited in the United States Government Life Insurance Fund. The remainder were issued in June 1937.
11This fact leads me to stress at this point that it is actual spending that has to be examined and nothing else and not any fancy constructs designed to rationalise government spending as the cure for recessions. Now it is argued that the net government contribution to spending dropped by $3.3 billion dollars and that this contraction was responsible for the 1937-38 crash. Unfortunately some people on the net think the preceding figure is the definite amount by which government spending fell: it is nothing of the kind.
12Unlike America there was little work-sharing in Australia and none at all in manufacturing, including reduced working hours. This makes the American recovery look even worse in comparison because reduced working hours and work-sharing is a form of hidden unemployment.
13Professor Sinclair Davidson and Julie Novak both made this mistake in their Five and a half big things Kevin Rudd doesn’t understand about the Australian economy. They also made the additional error of stating that Australia left the gold standard in 1931. Australia was actually on a sterling exchange standard that it abandoned in December 1929. Professor Davidson repeated these errors in his The Stimulus and the great recession.
Sinclair Davidson is Professor of Institutional Economics in the School of Economics, Finance and Marketing at RMIT University and a senior fellow at the Institute of Public Affairs. Julie Novak is also a fellow at the Institute of Public Affairs.
14Net government income creating expenditures are those that are spent on current output.
15Schedvin’s table only went to 1936. I extended his table by using the same data from the Official Yearbook of the Commonwealth of Australia.
16This refutes the absurd argument of some so-called economists that there is no relation between excessive wage rates and unemployment
17Working hours in Australian manufacturing never fell below 44. In short, there was no hidden unemployment.
18Statistical Abstract of the United States 1939. I think it is very important to continue to stress the fact that during the whole of the Great Depression the working week in Australian manufacturing never fell below 44 hours, meaning that there was no hidden unemployment. If America’s unemployment rate had been adjusted according to hours worked it would have been significantly higher.
19I ignored a direct cut in money wage rates to make the adjustment because the current state of public opinion would probably not support this measure. The best approach is to educate the public about the danger of excessive wage rates.
20Henry Thornton noted that inflation increases the demand for labour by lowering the cost of labour. He also drew attention to the consequences of sticky wages during a deflation. (Henry Thornton, An Enquiry into the Nature and Effects of the Paper Credit of Great Britain, 1802, Augustus M. Kelley, New York 1965, pp. 118, 189-90). Robert Torrens made the same observation in An Essay on the Production of Wealth (Longman, Hurst, Rees, Orme, and Brown, Paternoster Row, 1821, pp. 326-27). These men fully understood that inflation restored price margins and increased the demand for labour by raising prices.
21Today’s economists dismiss this fact on the absurd grounds that it is double-counting. Only the Austrians realised the vital importance of aggregate spending to the economy. In fact, it is impossible to properly discuss the concept of a production structure in the absence of aggregate spending. This Austrian insight has now been validated by the Bureau of Economic Analysis that has started to use a gross output concept (aggregate spending) to measure the economy for each quarter. This will finally confirm the Austrian argument that consumption does not drive the economy. (The Wall Street Journal, Mark Skousen, At Last, a Better Economic Measure).
22Writing in 1871 Stanley Jevons unleashed a strident attack on all four of Mill’s propositions on capital, reserving most of his ire for the fourth one. The extraordinary thing is that Jevons inadvertently made it clear that he had absolutely no understanding of what Mill had really said. Apparently following in Jevons’ footsteps Simon Newcomb, a prominent American economist, also launched a detailed attack on Mill’s fourth proposition. He too had failed to comprehend the reasoning behind Mill’s position. Edwin Cannan was another eminent economist who attacked the fourth proposition. Like Jevons he too failed to grasp what Mill really meant.
(W. Stanley Jevons, The Principles of Economics, London: Macmillan and Co., Limited, 1905, Ch. XXIV. Simon Newcomb, Principles of Political Economy, New York: Harper & Brothers, 1886, pp, 434-40. Edwin Cannan, The History of the Theories of Production and Distribution from 1776 to 1848, Staples Press, 1953, pp. 379-81).
If these critics had paid closer attention to Mill they might have detected the influence of John Rae’s views on capital. It was Rae’s brilliant insights into the nature of capital that directly influenced the thinking of Nassau Senior and Mill on the subject, a fact that they both admitted. That Mill’s critics were unaware of John Rae’s original contribution to capital theory is made clear by the fact that their books did not contain a single reference to the man. It is apparent from Jevons’ comment regarding Jeremy Bentham’s opinions on capital that he had no knowledge of Rae’s work.
(John Rae, The New Principles of Political Economy, 1834, reprinted by August M. Kelley, New York, 1964).