An important argument in the arsenal of Keynesian economics in favor of fiscal stimulus is that during a recession/depression it will put unemployed resources back to work and produce, in the process, a net-gain in wealth.
Keynesians begin with a fairly accurate description of a few crucial facts. Immediately before and during a crisis, fully equipped factories, tools, materials, implements, workers able and willing to work, materials, the know-how etc. are all in place and ready to be used. Yet, during a crisis the only thing that seems to be lacking is enough spending to facilitate the exchange of goods and services. Throughout the economic system virtually every industry, consumers’ and capital goods’ alike, is lacking in money demand for its products. Capital goods and workers are idle in bakeries, in shoe factories, in steel companies, in the construction business, in mining concerns, transportation industries etc. etc.
Closely related, typical depression conditions, particularly at the outset of a contraction, are characterized by abnormally large inventories of previously produced physical goods. While idle, stocks of consumers’ as well as capital goods of standard quality and usefulness are losing rapidly in quality and value. It is a well-established empirical fact that the so-called physical subsistence fund is not at all low or severely depleted but, on the contrary, reaches its maximum level precisely at the zenith of a boom. The inventories of perishable goods, such as food, are in most urgent need to be put to use as quickly as possible to prevent consumption of that capital. To avoid capital losses, sellers are eager to sell such money-losing, excessive inventories and regain the condition of normal operation and production, but there are not enough buyers.
On this point, some Keynesians insist emphatically, and I believe correctly, that from the standpoint of physical production the “natural” process of liquidation of inventories will only aggravate the crisis by extinguishing stocks of physical goods which are so desperately needed for the continuation of production and further capital accumulation. Malinvestment is a very serious problem but its negative consequences are less severe and will be corrected much faster once all productive resources are put back to work.
To revive the economic system, to restore spending, they conclude, a considerable push in effective aggregate demand is called for.
I submit that as far as these facts of depression are concerned there is nothing at all that is objectionable. Furthermore, an increase in aggregate spending is indeed potentially capable to reemploy idle men and machines and close the output gap.
There is no doubt that the observed sharp decline in business profitability did in fact originate on the side of money and spending and not on the side of physical production, worker preferences for leisure, expectations and the like. Consequently, there is also little doubt that a recovery in business profitability, one way other another, will be brought about by an adjustment of money cost of production to money sales revenues, which at the present are abnormally large.
To have a recovery, either of one of two or both things must happen: the aggregate amount of money cost of production will have to fall or the aggregate volume of money sales revenues will have to rise. Ideally, in an environment of a severe financial contraction, such as we have now, unrestricted freedom of competition in labor and product markets will bring down money cost of production quite rapidly by means of swift and deep cuts in nominal wages and capital goods’ prices. Thus, the most efficient path to recovery is through reductions in nominal wages and prices.
But, and this is also important to realize, the mismatch between the volume of production, employment and the corresponding absolute heights of money prices and wages, on the one hand, and volume of spending and revenues, on the other, is potentially capable of a successful correction by means of a sufficient increase in the aggregate volume of spending. In principle, the two methods are equally potent in overcoming adverse consequences of a financial contraction.
But, an attentive reader might ask, does not this latter point actually concede one of the key Keynesian pieces of analysis, namely, that what matters is sufficient effective aggregate demand? By no stretch of imagination! As I will attempt to demonstrate below, the actual key question in the debate is not about more or less spending but which kind of spending.
Keynesians, by virtue of the specific structure of the Keynesian system, see in consumption and government expenditure the remedy for inadequate aggregate demand. Furthermore, in this framework the problem fundamentally responsible for the inadequacy of aggregate demand is excessive savings.
By contrast, a consistent Austrian analysis sees the solution in more saving as the foundation for more productive expenditure. A consistent Austrian sees in excessive private and public consumption, particularly public consumption, the most significant obstacle to recovery. The issue is the distribution of spending between consumptive and productive.
To highlight the differences between Austrian and Keynesian understanding of the interdependence between production/employment and spending, I have constructed two hypothetical scenarios each involving a distinctive expenditure-augmenting government stimulus package to explicitly address the problem of reduced aggregate demand in an environment of sticky wages and prices. Scenario A (Stimulus Package A) makes use of the Austrian concept of structure of production and the vital distinction between productive and consumptive spending. Scenario K (Stimulus Package K) follows closely the fundamental mechanics of Keynesian economic analysis which emphasizes consumption spending as the determinative factor in the structure of spending.
To be sure, only Austrian scenario is purely hypothetical. As pointed out above, Austrians understand the free-play of competition as the most potent means to overcome particularly the short-run mismatch between an excessive boom-level of nominal wages/prices and depressed crisis-level volume of aggregate spending. The stimulus Package K, on the other hand, in all its essential features is already in place which, quite conveniently, renders its analysis only much more realistic and relevant.
The thought experiment is designed to provide the reader with a direct comparison of major analytical claims of the two competing approaches as well as a framework to assess the ability of each of the two to affect, positively or negatively, employment, capital accumulation, and the general standard of living/real wages.
Scenario A—The Austrian Stimulus Package
Assume the government agrees that all branches of the economy, although to a different degree, suffer from a falloff of aggregate demand. In response, the government invites a group of Austrian economists to formulate a scheme in accordance with the concept of capital structure. Austrians organize their package around the cornerstone of their macroeconomic theory—the synchronized character of production of a modern economy consisting of a multitude of simultaneously operating stages of production.1
Austrian and Keynesian economists agree about the size of the additional $700 billion to boost the aggregate demand. They do disagree, however, and disagree vehemently, about the targeted macroeconomic aggregates—precisely the main issue here. Now, the government drafts and ratifies the Austrian inspired Capital Structure Recovery Act (CSRA) to the tune of the said amount, obtains the newly printed cash from the Fed and distributes these funds to companies and consumers in the economy.2
Austrians explain that in order to achieve maximum benefit, the solution cannot be simply to spend the CSRA money indiscriminately but to place the funds in such a way as to approximate the qualitative and quantitative structure of spending under normal, neither inflationary nor deflationary, conditions of production. And under normal conditions, Austrians insist, the relative quantitative importance of productive expenditure vis-à-vis consumption spending is overwhelming.
They argue that while consumptive expenditures out of current wage income or dividend and interest payments constitute the sources of funds to the sellers of consumers’ goods, the spending for capital goods and labor in all other stages down the chain of production must come from funds that are saved and productively expended, not spent on consumers’ goods. Moreover, the size of aggregate consumer spending in any given period is by several orders of magnitude less than productive spending based on gross savings of businessmen and capitalists.
According to the parameters of the Austrian package, the CSRA money enters the economy not only, not even primarily, through the retail sector (consumer spending) but also, and overwhelmingly, through the capital goods’ producing industries (productive spending). Businesses along the entire synchronized structure of production are eager to exchange some of their inventories against newly created cash. Very importantly, from the previous boom inherited excessive inventories do not simply perish but put to work quickly and contribute greatly to recovery and further capital accumulation.
Particularly, businesses in the higher stages of production (capital goods industries), who suffer the most during the downturn, benefit from the funds that enable them to stay in business and supply companies down the structure of production with necessary inputs. Machines and workers are busy producing again, sending the finished and semi-finished goods up and down the entire structure of production. Both the physical and financial aspects of economic activity are fully geared towards reaching full-employment equilibrium and increased production.
Applying the logical structure of their analysis further, Austrians propose to eliminate capital gains taxes, taxes on profits, inheritance taxes and all other taxes that predominantly fall on capital incomes and incomes of wealthy and super-wealthy. The resulting shortfall of tax revenues should be greeted as a welcome opportunity to slash government spending. Austrians motivate their proposal on the ground that those taxes reduce the productive expenditure relative to consumptive expenditure and thereby needlessly delay and take away funds from reemployment and wealth creation.3
Rather than aggravating economic inequality the proposed tax cuts, Austrians emphasize, will most likely not be used for personal consumption but spent as additional productive expenditure directly increasing the demand for labor and capital goods. They particularly stress the fact that it is businesses, not consumers, who make wage payments, purchase capital goods, establish and improve distributional mechanisms, raising thereby productivity of labor and real wages. Businesses, not consumers, are the pillars of a modern economy, the productive engines that organize production and offer wages, and if left free to save and accumulate capital are virtually compelled by the forces of economic competition to continuously increase the quantity and improve the quality of goods for the benefit of wage earners.
The policy relevant bottom-line is that if governments are serious about economic recovery, rapid reemployment and improving productivity of labor, they should pay close attention to the characteristic elements of an economy based on a complex structure of production and division of payments with saving and productive expenditure being its most vital elements.
The fundamental theoretical insight is that a positive marginal effect on employment and production from an upward push of aggregate demand (financed by monetary expansion) is positively and linearly correlated with saving and productive expenditure, and negatively and linearly with consumption expenditure. To obtain a maximum economy-wide positive impact on the demand for labor and capital goods, a maximum proportion of CSRA funds must be saved and productively expended.
Scenario K—The Keynesian Stimulus Package
Eagerly agreeing to the amount of the funds needed and emphasizing furthermore that rigid wages and prices will actually help the economy to recover, a group of Keynesian economists presents its stimulus package called Maximum Consumption Revival Act (MCRA), drafted in full accordance with the conceptual framework and theoretical content of Keynesian/Neoclassical Synthesis economics. The draft emphasizes to boost consumption spending, preferably government spending on a variety of assorted programs. The proposal for greater consumption is given the following theoretical support.
What explains all sudden reductions in aggregate demand, we read in the draft, is a fundamental and systematic disequilibrium between intended saving and intended investment. In the present crisis the economy was brought to its knees by a disastrous combination of a global savings glut and diminished investment opportunities.4
Currently, the mismatch has become so enormous that the interest rates are near their lowest boundary, gripping the economy in the liquidity trap. Since it would amount to a crime continuing feeding the economy with the same poison, it is of supreme importance to divert those idle savings back into the spending stream by means of increased government consumption, i.e. fiscal policy.5 Consumption constitutes 70 percent of the national income, thus it is a good proxy to assume that 70 percent of wages and all other incomes are paid by consumption expenditure, the rest being paid by investment.
The primary objective is to ensure that the package is devised in such a manner as to prevent any possible disastrous leakage into saving along the way. There are a number of proposals to reach the maximum level of consumption.
One such measure to prevent MCRA money from being saved and reinvested is to actually enforce the absolute minimum of socially tolerable level of saving. Anyone who is eligible to receive MCRA money will be required to provide an ex-post proof that he, in fact, used at least 70 percent of his income, in accordance with the average propensity to consume of approximately 70 percent, for further consumption. (Actually, in view of the global saving glut and liquidity trap, his consumption share should be much greater than 70 percent for there are simply not enough investment opportunities for the whole of 30 percent of income saved to be profitably absorbed!)
For example, if it can be proven that a grocery store owner after having sold $100 worth of goods did not consume at least $70 but instead followed his usual practice of saving and productively expending $90 for (1) paying his employees ($40), (2) purchasing goods from wholesale ($40), (3) paying the heating/lighting and other business related bills ($10), he will be fined to the full equivalent of funds thereby saved and productively expended. The measure would greatly aid in the establishment of a socially conform incentive structure; suppress selfish and socially destructive phenomenon of excessive saving.
Certainly, to some the measure might appear draconian but it is clearly a lesser evil if contrasted with disastrous consequences associated with saving. In accordance with the exacting standards of the public choice theory, a carefully designed cost-benefit analysis shows beyond doubt enormous social gains from a stricter control of saving behavior. First of all, in view of an already established and quite effective system of tax collection and enforcement the social costs in connection with monitoring and enforcing should be minimal.
More importantly, if contrasted with the enormous financial losses suffered by millions upon millions of households that were ultimately caused by the savings glut such surgical and effective government intervention would produce a considerable net-social benefit. The negative externalities of an excessive propensity to save will be fully internalized to prevent any future build-up of excessive savings—the sum of all fears, threatening global economic stability.
An alternative, slightly more bureaucratic and costly, if not less intrusive, measure would be to step up the progressivity of the federal income tax, impose a heavy marginal tax on capital gains and inheritances. Since what brought about the crisis was a combination of an unprecedented lack of investment opportunities/savings glut, the $700 billion stimulus package, however favorably towards stepped-up consumption, may actually be not enough.
In view of the enormous gap in income and wealth in American society, there is ample room for macroeconomic stabilization, Keynesian economists insist. The problem is that rich people save more absolutely and most likely also relatively, i.e. they have a higher propensity to save. On the other hand, people with lower incomes have a much lower propensity consume. A policy of redistribution of purchasing power would ensure a pattern of spending that would forestall any further increases in saving. Nobel Laureate in economics, Prof. Joseph Stiglitz explains:
Within the sphere of changes to taxes and transfer programs, the impact on the economy depends primarily on the propensity to consume -- that is, on how much of an additional dollar of income is spent rather than saved -- among those who receive the transfer payments or pay the taxes. The more that the tax increases or transfer reductions are focused on those with lower propensities to consume (that is, on those who spend less and save more of each additional dollar of income), the less damage is done to the weakened economy. Since higher-income families tend to have lower propensities to consume than lower-income families, the least damaging approach in the short run involves tax increases concentrated on higher-income families. Reductions in transfer payments to lower-income families would generally be more harmful to the economy than increases in taxes on higher-income families, since lower-income families are more likely to spend any additional income than higher-income families. Indeed, since the recipients of transfer payments typically spend virtually their entire income, the negative impact of reductions in transfer payments is likely to be nearly as great as a reduction in direct spending on goods and services.6
The policy relevant bottom-line is for governments to recognize that the ultimate cause of economic depressions in advanced capitalist economies is a chronic tendency of savings to outstrip investment, thus creating the problem of insufficient effective aggregate demand. The political challenge is to devise a counter-cyclical program that should include automatic stabilizers (induced taxes and transfer programs), expansionary fiscal policy (budget deficits) aimed at using excessive/idle private saving for public sector investment, progressive taxation to prevent further accumulation of savings, and an aggressive monetary policy of credit expansion to lower interest rates.
The fundamental theoretical insight is that a positive marginal effect on employment and production from an upward push of aggregate demand (financed by monetary expansion) is positively and linearly correlated with consumption expenditure. Current employment and production are stimulated directly and immediately if there is an increase in consumption expenditure. Certainly in the short-run, the employment and output effects are the greater the greater is the marginal propensity to consume.
From the standpoint of the impact on the demand for labor and capital goods, consumption and saving/investment are indistinguishable in their effect. One will raise the demand for labor and maintain existing production structure equally successfully by means of both consumption and investment spending. Net-saving/net-investment out of income is important but only as means for the production of additional capital goods. Since the fundamental cause of recessions/depressions is a disastrous lack of effective demand, caused primarily by an excess of saving, stepped up private consumption and public “investment” at the expense of saving “provides much more bang for the buck.”7
In Lieu of Conclusion
An Austrian stimulus package is, of course, the only fictional idea in the reported thought experiment. An “Austrian recovery” does not depend on any outside, i.e. governmental, stimulus, however depressed the general economic conditions may be. In the Austrian view, to restore production and return to full-employment, neither the spending of additional funds nor the redistribution of the purchasing power from the private to public sector is required.
Practically, all one need to do is not to interfere with the fundamental market forces of supply and demand but let them their job. In the face of a lower overall volume of spending (which, incidentally, was brought about by the preceding credit expansion aided and encouraged by central banks, i.e. by the policy of interventionism), the self-interest of buyers and sellers of goods and services is the most reliable mechanism to adjust prices and wages to the new reality of a lower aggregate monetary demand in the economic system.
Analytically, the key challenge is to understand that consumptive spending is not a substitute for productive spending. At the most fundamental conceptual level, when it comes to sustaining the existing capital/production structure, a sharp line must be drawn between the two mutually exclusive categories of spending. It is the production spending by businesses that maintains the production structure and supplies the funds to buy capital goods and employ workers. Consumptive spending is important but it contributes neither to production nor to more employment. Indeed, its effect is actually the opposite. A higher relative spending for consumers’ goods works against economic recovery.
The contemporary mainstream macroeconomics profession, of both Keynesian and neoclassical variety, commits the fundamental error by equating consumer spending by anyone, for any purpose, with the demand for productive resources (capital goods and labor services). Their understanding, at the bottom, is that what finances the employment of productive resources is simply spending. Consequently, the problem of aggregate demand is that there is a lack of spending, of spending of any kind. No distinction is made between consumptive and productive spending.
To illustrate this, let us ask a very simply question. What is it that pays wages and determines the level of employment (with a given supply of labor)?
The Keynesian answer is that it is the effective demand that determines it which, in turn, consists of (a) consumer spending and (b) investment spending.
Observe that when it comes to maintaining the level of wages fundamentally the internal composition of the aggregate demand, i.e. the respective shares of consumption and investment, is of no consequence at all.
Investment spending bears the burden of an outlay for household’s net-saving, and insofar net-saving is over and above the depreciation charges, it constitutes net-investment and contributes to the “real” growth of the economy. This is the only role of investment spending as conceived by the Keynesians/neoclassical economists.
Note the crucial condition which is implied in this theory.
Zero net-saving (out of income), and thus zero net-investment, imply and are perfectly compatible with full employment. That is, if all households (workers, businesspeople, capitalists) decided to consume all of their incomes, the only conceivable loss would be a foregone increment of the (future) national income. Under such a condition, according to Keynesians, there is absolutely nothing that would threaten business profitability and employment.
Logically then, troubles can only come from things that induce a certain portion of the aggregate spending to “leak” out of the system. Consequently, those who attempt to analyze economic phenomena using Keynesian framework as analytical device will look for those “leakages” to explain business failures and unemployment.
Closely related, those who attempt to criticize some of Keynesian elements in theory and policy but accept its fundament framework will be driven to look for mechanisms to fix the “leakages” via interest rates, studies of behavioral parameters determining consumption function and the like. The crucial thing is that they are in agreement with the fundamental premise of the Keynesian framework.
It should be understood that the greatest blunder of the entire Keynesian system is not merely that it views saving, domestic or global, as “leakage” that under certain circumstances causes desired savings to exceed planned investment. No, the much more serious error is that it does not recognize the fact that it is saving—first and foremost the gross saving by businessmen and capitalists out of sales revenues—which constitutes the very financial (monetary) source and means for the overwhelming share of aggregate spending that goes on in an advanced market economy. None of this is acknowledged, either explicitly or implicitly, in the Keynesian account of things. To the contrary, it is openly contradicted by the very mechanics of Keynesian theoretical system.
To put it another way, Keynesian analytics is completely oblivious to the fundamental distinction between the two mutually exclusive categories of productive and unproductive (consumptive) spending. That there even exist such a distinction and that is so decisive precisely when it comes to the level of employment is not, and cannot, be accounted for in Keynesian analytics. For Keynesians, the fundamental problem is to get people spend for anything at all—and that spending as such, without an explicit distinction between categories of spending—which pays wages, purchases and replaces capital goods, and ensures business profitability. But this view is squarely at odds with economic reality.
One of the key insights of Austrian economics is that in a modern division-of-labor economy, the decision to produce or abstain from production of a good in question is ultimately decided by the specific structure of consumer spending. But the structure of consumer spending is most certainly not the end of the story.
Consumer spending qua consumer spending does not determine whether more or less consumers’ goods as a whole relative to capital goods as a whole will be produced. Indeed, if we assume a fixed aggregate volume of spending in the economic system, an increase in consumer spending can only come at the expense of less spending for labor and/or capital goods.
Just as the relative strength of the demand for consumers’ goods A vs. B determines the relative profitability of employing existing factors of production in the production of A or B, exactly the same mechanism, but at a higher level of aggregation, determines the relative profitability of employing existing factors of production (labor and capital goods) in the production of consumers’ goods, as a broad category, or in the production of capital goods, as another broad category.
The choice, from the standpoint of the economic system as a whole, is not only to decide whether to produce A or B but also whether existing factors of production are employed in the production of consumers’ goods or in the production of capital goods.
If the entire revenue of the economy would be spent on consumers’ goods, round after round, then the only kind of spending present would be the spending that circulates within the consumers’ goods sector only. Essentially, the money would go from A to B and back again. In the process, not so much as a single cent would trickle down from retailers to their employees as wages, nor to supplying industries and their workers and suppliers etc., down the entire chain of production and distribution. Thus, with no fresh goods coming from suppliers, businesses down the production structure would eventually run out of their inventories and leave the few remaining consumers with nothing to buy.
Fortunately for the welfare of us all, an actual economy does not at all fit the description of Keynesian mechanics.
Real world entrepreneurs, throughout the entire production structure, are rational enough to anticipate and meet the demand by saving and productively expending their revenues. If they are not prevented from saving and preserving their capitals, capital goods and consumers’ goods are produced and people are employed and paid wages. The production structure is maintained because of a simple calculation that if every single cent they earned were spent on present consumption, there would simply be no means to keep the businesses going.
Only when people were coerced through taxation and/or inflation to spend their entire incomes and revenues on present consumption, the economic system would work along the Keynesian scenario, with the most disastrous consequences for production and employment.
- 1For the most advanced qualitative and qualitative elaboration of the concept of structure of production, see George Reisman (1996), Capitalism: A Treatise on Economics, chapter 17, especially pp. 838–59. For a more formal exposition, see Renaud Fillieule, “A Formal Model in Hayekian Macroeconomics: The Proportional Goods-in-Process Structure of Production,” Quarterly Journal of Austrian Economics (2007), vol. 10, pp. 193–208. www.mises.org/journals/qjae/pdf/qjae10_3_1.pdf.
- 2The essence of the scenario is very similar to the proposal of Prof. Reisman to establish a 100 percent reserve gold standard. See George Reisman, (unpublished paper, March 25, 2000), George Reisman, “The Goal of Monetary Reform;" idem, “The Path to Sound Money” (audio recording, Aug. 4, 2007). The key idea in his proposal is to shore up the balance sheet of commercial banks through an infusion of a sufficiently high priced stock of gold. The proposal would serve to maintain the volume of spending at an appropriately high level, comparable to that before the crisis, with the added advantage that it will eliminate the fractional reserve banking by introducing the important element of 100 percent specie reserve. If the goal is to avoid a disastrous depression and to introduce a vital element that would prevent all further boom-bust episodes from happening, the introduction of the 100 percent gold standard as proposed by Prof. Reisman is an even much better scenario for dealing with the problem of economic crisis. Nevertheless, for the sake of illustration of the essence of the problem, Scenario A is a reasonably good approximation.
- 3For an elaboration, see George Reisman, “For Society to Thrive the Rich Must Be Left Alone,” Mises Daily (March 2, 2006), http://mises.org/story/2073.
- 4In his appropriately titled Op-Ed column “Revenge of the Glut,” Prof. Paul Krugman concludes: “[o]ne way to look at the international situation right now is that we’re suffering from a global paradox of thrift: around the world, desired saving exceeds the amount businesses are willing to invest. And the result is a global slump that leaves everyone worse off. So that’s how we got into this mess. And we’re still looking for the way out.” Paul Krugman, “Revenge of the Glut,” New York Times (March 1, 2009), http://www.nytimes.com/2009/03/02/opinion/02krugman.html?_r=0
- 5See Christina Romer & Jared Bernstein, “The Job Impact of the American Recovery and Reinvestment Plan” (2009) http://otrans.3cdn.net/45593e8ecbd339d074_l3m6bt1te.pdf. The authors write: [t]he indirect effects are those coming from the fact that the newly employed workers spend more and this stimulates other industries. For core spending programs, we assume the direct output effects move one-for-one with the spending increase. Broad tax cuts have jobs effects, but they stem only from indirect effects: tax cuts only have effects when people go out and spend the money… It is important to note that the jobs effects of temporary broad-based tax cuts would probably be considerably smaller. Large proportions of temporary tax cuts are saved, blunting their stimulatory impact on output and employment. [p. 6]
- 6Peter Orszag & Joseph Stiglitz, “Spending Cuts vs. Tax Increases at the State Level: Is One More Counter-Productive than the Other During a Recession?” (Center on Budget and Policy Priorities, Oct. 31, 2001), www.jacksonprogressive.com/issues/econandwelfare/statetaxes110201.html.
- 7Cf. Paul Krugman, “Bad Faith Economics,” New York Times, January 26, 2009, p. A23.
