Keynes’ Long-period Problem of Saving

Keynes did not present his general theory as a static equilibrium system embodied in a set of simultaneous equations. That was done by Hicks, Hansen, Samuelson, Patinkin, and countless other Keynesians. Their approach was Walrasian, and as a result of their efforts a Walrasian revolution took place in economics following Keynes’ publication of The General Theory of Employment, Interest, and Money. That revolution took Keynes’ name in spite of the fact that Keynes was a protégé of Marshall, not Walras, and Keynes emphatically rejected the kind of mathematical equilibrium approach embraced by the Walrasians that ignore the problems imposed by empirical reality. [Blackford (2016a)]

Keynes spent his entire life applying Marshall’s methodology to the analysis of economic problems in an attempt to discover the causal relationships that drive the economic system through time. [Blackford (2016a)] In so doing, among many other things, it became clear to him that:

1. The ultimate justification for production is to satisfy the demands of consumers.

2. The true causal variables in a market economy in which the processes of production take time are expectations with regard to the future.

3. Employment, output, and income are determined by effective demand, that is, by the demand that producers expect to prevail when the goods they are currently producing are to be offered for sale.

4. Rates of interest are purely monetary phenomena determined by the supply and demand for liquidity as the prices of assets adjust to equate wealth-holders’ demands for assets to the existing stocks of assets.

5. The willingness to invest in the production of newly produced real assets depends on the prospective (expected) return on investing in those assets.

6. The prospective return on investing in real assets is affected not only by rates of interest (which affect the cost of financing investment) but also by the stocks of existing real assets relative to the demands for the outputs those real assets produce (which affect the expectations with regard to the revenues that can be obtained from increases in the stocks of those assets).

7. Monetary policy is limited in its ability to affect rates of interest by the propensities of wealth holders with regard to their demands for liquidity.

8. Once rates of interest have reached the lower bounds set by the propensities of wealth holders the level of economic activity is determined by the effective (expected) demands for consumption and investment goods.

9. Since the ultimate justification for production is to satisfy the demands of consumers, given the existing stocks of real assets and rates of interest, the demand for investment goods is ultimately determined by expectations with regard to future consumption.

10. And since expectations with regard to future consumption are largely determined by current consumption, the demands for consumption and investment goods (hence, the level of economic activity) must be largely determined by current consumption and, ultimately, by the way in which current consumption, rates of interest, and the existing stocks of real assets change over time. [Blackford (2016b)]

Even though these assumptions and arguments stand at the very core of Keynes’ general theory, this way of looking at the economic system was not taught or even acknowledged in graduate schools when I was in academia, and I suspect it is not taught or acknowledged in graduate schools to this day. Instead, graduate students were taught the Walrasian paradigm with a make-believe tâtonnement/ recontract dynamics (which assumes no transactions take place until all markets are in equilibrium) while the Marshallian paradigm from which the dynamics of Keynes is drawn was pretty much ignored other than in undergraduate courses. [Blackford (2016a)]

At the same time economists have invested an incredible amount of time and energy over the years debating what Keynes was actually trying to say in The General Theory as they ignored Marshall and attempted to force Keynes into a Walrasian mold. This is especially bizarre in view of the fact that virtually nothing Keynes actually said in The General Theory makes sense when viewed from the perspective of the Walrasian paradigm and can only be understood from the perspective of Marshall.

Even more problematic is the fact that this has led to a situation in which mainstream economists have failed to appreciate the dynamic nature of causality in Keynes’ analysis and, in particular, the implications of Keynes’ analysis as to the way in which a market system of economic organization must evolve over time. [Blackford (2016a and 2016b)] This failure has led to dire consequences, not only for the discipline of economics, but for the world at large in that it has made it impossible for mainstream economists to understand Keynes’ analysis of what Robertson dubbed “the long-period problem of saving.” It is argued below that it was this failure that led mainstream economists to support the economic policies implemented over the past forty years that are responsible for the economic, political, and social crises we see developing throughout the world today.

Robertson, Keynes, and the Long-Period Problem of Saving

Robertson argued that an increase in the propensity to save will lead to an increase in the supply of loanable funds (willingness to lend) that will cause the rate of interest to fall and that this, in turn, must lead to an increase in the willingness to purchase newly produce capital goods (i.e., investment). Thus, according to Robertson, it is reasonable to expect a smooth transfer of resources out of the consumption-goods industries and into the capital-goods industries as a result of an increase the propensity to save. He also believed the increase in the stocks of real capital assets that will result from the increase in saving will increase productivity and, therefore, economic growth and wellbeing in the future. Keynes did not see this sequence of events working itself out in this way at all.

Keynes saw this process as beginning with a ceteris paribus situation in which—given the expectations on which employment, output, and income depend—in the face of an increase in the propensity to save (which, by definition, means a decrease in the propensity to consume) businesses in the consumption-goods industries would no longer be able replenish the money balances needed to finance their transactions from sales. As a result, in order to meet their expenditure obligations they would be forced to increase their borrowing (i.e., increase in their demands for loanable funds) in order to obtain the requisite balances needed to maintain the level of employment and output that maximized their expected profits.

Thus, Keynes argued that, given the profit-maximizing expectations of businesses, there is no reason to expect the increase in the propensity to save to have a direct effect on the rate of interest as the increase in savers’ willingness to lend would be offset by the increase in business’ willingness to borrow. He further argued that, given the supply of money, it is only after expectations change with regard to the profitability of continuing to produce at the current level of employment and output that businesses would allow the level of employment, output, and income to fall in the consumption-goods industries and that it is only after employment, output, and income fell that the rate of interest would fall. [Blackford (2016b)]

Keynes’ argument, as stated above, assumes, of course, a fixed supply of money, and, as Robertson correctly pointed out, there is no reason to believe the interest rate will not fall simultaneously with an increase in the propensity to save in Keynes’ scenario if the increase in saving takes the form of an increase in the demand for securities and is accompanied by an increase in the supply of money. [Blackford (2016b)] Nevertheless, Robertson’s argument fails to address the fundamental issue raised by Keynes.

If you believe, as Keynes believed, that production takes time and that producers must be guided by their expectations with regard to the profitability of continuing to produce in the consumption-goods industries at the current levels of employment, output, and income there is no reason to believe employment, output, and income will begin to fall in the consumption-goods industries, no matter what happens to the rate of interest in this situation, until after there is a change in expectations with regard to the profitability of continuing to produce in the consumption-goods industries at the current levels of output, employment, and income.

Furthermore, if you also believe, as Keynes believed, that all production is ultimately for the purpose of satisfying the consumer, there is also no reason to believe that the subsequent fall in expectations in the consumption-goods industries with regard to the profitability of continuing to produce at the current level of employment, output, and income will not have an adverse effect on expectations with regard to the profitability of investing in those industries. This means that the resulting fall in the rate of interest that is supposed to increase investment and, thereby, increase the stocks of real assets is apt to be accompanied by a fall in expectations with regard to the profitability of investing in those industries.

In other words, Keynes argued that there is no a priori justification for assuming that, in the end, the net effect of the two opposing forces acting on investment will be an increase in the accumulation of capital in the long run. The positive effect on investment from the fall in the rate of interest could be more than offset by the negative effect from the change in expectations with regard to the profitability of investment. Thus, there is no a priori reason to believe that the increase in the propensity to save will ultimately lead to increase in the accumulation of capital in the long run as this dynamic process plays itself out. Keynes argued that an increase in the propensity to save (decrease in the propensity to consume) would most likely lead to a decrease in employment, output and income over time, rather than to an increase in the accumulation of capital, and that this result is a virtual certainty if the increase-in-saving/fall-in-consumption were to turn out to be permanent as expectations adjusted to this reality in the long-run. [Blackford (2016b)]

Unfortunately, this kind of analysis has proved to be beyond the grasp of mainstream economists.

The Great Recession of 2007 - . . .

As a result of the inability of mainstream economists to grasp the importance of the relationships between employment, output, income, and expectations explained by Keynes in The General Theory, mainstream economists have failed to grasp the simple point that Keynes attempted to drive home throughout The General Theory, namely, that economic growth in the long run is enhanced by an increasing propensity to consume and is inhibited by an increasing propensity to save.

As a result, the economic models created by mainstream economists over the past fifty years have ignored the long-run relationship between consumption and effective demand. These same models have been used to justify deregulating the domestic and international financial systems, cutting corporate taxes and taxes on the wealthy, increasing taxes on the not so wealthy, reducing public investment in infrastructure and human capital, eliminating usury laws, destroying labor unions, promoting the adoption of private retirement accounts, converting Social Security from a pay-as-you-go to a partial-prepayment system, neglecting the minimum wage, and many other supposedly economically efficient policies that only make sense within models that ignore the long-run relationship between consumption and effective demand and assume that increasing saving enhances economic growth. [Blackford (2014, Ch 1; 2016b)]

These policies have led to a dramatic increase in saving in the foreign sector (by way of our increased current account deficit) and at the top of the income distribution in the private sector (by way of the increase in the concentration of income at the top of the income distribution) over the past thirty-five years. These increases in saving were partially offset by dissaving in the public sector and at the bottom of the income distribution in the private sector leading up to 2007. They were also accompanied by increases in investment as a result of speculative bubbles in the commercial real estate markets in the 1980s, in the markets for tech stocks and the dotcom and telecom industries in the 1990s, and in the housing market in the 2000s. The problem is that the only way in which we were able to achieve full employment during this period was through an increase in debt relative to income. [Blackford (2014, Ch 1, Ch 2, and Ch 3)]

Herein lies the fundamental problem we face today: The economic policies we have followed since the 1970s have created a situation in which—given a) the increased concentration of income, b) the degree of mass-production technology that exists within our economic system, and c) the size of our current account deficits—it is impossible to sustain the mass markets (i.e., the large numbers of people with purchasing power) needed to fully employ our economic resources in the absence of a continual increase in debt relative to income. [Blackford (2014)]

This is a problem because continually increasing non-federal debt relative to income is unsustainable in the long run since it increases the transfer burden on debtors as income is transferred from debtors to creditors through the payment of interest. This means that eventually the system must breakdown as interest payments increase and non-federal debtors eventually find it impossible to meet their financial obligations. [Minsky] This leads to financial crises in which the financial system must be bailed out by the federal government to keep it from collapsing and bringing the rest of the economic system down with it.

By the same token, continually increasing federal debt relative to income is a problem because as interest payments on the federal debt grow they must eventually overwhelm the federal budget. This will make it more and more difficult to fund essential government programs such as Social Security, Medicare, Medicaid, and national defense. And most important, an increase in government debt is likely to contribute toward a further increase the concentration of income. This will make the long-run problem we face even worse because a further increase in the concentration of income will increase the degree to which debt must increase relative to income in order to fully employ our economic resources. [Blackford (2014, Ch12)]

It is the unsustainability of a continual increase in private-sector debt relative to income that eventually led to the Crash of 2008, and it is the inability to further increase debt relative to income through dissaving at the bottom of the income distribution (or through continually increasing debt relative to income in the public sector) combined with a the lack of speculative bubbles to stimulate investment (and an inability to continual decrease in the trade deficit) that has led to the diminished long-term expectation with regard to consumption that is the primal cause of the economic stagnation we have experienced since 2007. [Blackford (2014, Ch 3 and Ch12; 2016b)]

In other words, what we are facing today is the fallout from Keynes’ long-period problem of saving which stands at the very core of Keynes’ general theory. The logic of his argument is this:

1. The accumulation of real capital assets over time reduces the demands for newly produced real capital assets (i.e., the investment demand schedule) by reducing the prospective yield on additional units of real capital assets at any given rate of interest as those assets become more plentiful or even redundant.

2. The failure of the propensity to save to fall (i.e., the propensity to consume to increase) over time at a rate that will offset the effects of the increase in the stocks of real capital assets on the demands for newly produced real capital assets (due to the fall in expectations with regard to the revenues that can be obtained from increases in the stocks of these assets) as these assets accumulate leads to a situation in which a fall in rates of interest is required (to lower the cost of financing newly produce real capital assets) in order to sustain the level of investment needed to maintain full employment.

3. Since rates of interest are determined by the supply and demand for liquidity as the prices of assets adjust to equate wealth-holders’ demands for assets to the existing stocks of assets, there are limits to the rate and extent to which rates of interest can fall. Thus, the failure of the propensity to save to fall (propensity to consume to increase) over time means that, to the extent the fall in rates of interest lag behind the fall in the demands for capital assets we can expect the economy to hit the lower bound set by the rates of interest. This, in turn, must cause investment to fall below the level needed to maintain full employment.

4. Even if interest rates were to adjust rapidly enough to avoid sporadic unemployment as capital accumulates in the short run, in the long run interest rates must eventually be forced to zero (after allowing for risk) and can fall no more.

5. Thus, the only way the economic system can be kept going in the short run is through a continual decrease in the propensity to save (increase in the propensity to consume) or fall in the rates of interest, and even then, as the system evolves over time, eventually rates of interest (after allowing for risk) and net saving and investment must be forced to zero as the system reaches a point at which so many real capital assets have been produced that additional increases in these assets will yield a negative prospective yield, and there is no reason to assume that the economy will be able to achieve full employment when this point is reached. [Blackford (2014, Ch12 ; 2016b)]

These ideas are not incorporated within the worldview of mainstream economists because the effects of capital accumulation and consumption on the prospective yield of increases in the stock of capital are not incorporated into the Walrasian models through which mainstream economists view the economic system. As a result, these models present a vision of the economic system in which monetary policy can be used to maintain full employment through stimulating investment indefinitely into the future simply by reducing rates of interest. This is a vision that presupposes monetary policy can be effective in maintaining full employment even if the entire country were to become paved over with concrete and every square inch of land were to sport a factory or high-rise apartment building. It also presupposes that the-more-we-save/the-less-we-consume the sooner this end can be achieved.

Keynes did not share this vision of the long-period problem of saving. As was noted above, Keynes believed that the only way the economic system can be kept going in the short run is through a continual increase in the propensity to save (decrease in the propensity to consume) or a continual decrease in rates of interest, and, even then, net saving and investment must eventually be forced to zero when there is so much capital that additional increases would be seen as having a negative prospective yield. [Blackford (2016b)]

The Crash of 2008

Even though Keynes analyzed the long-period problem of saving in excruciating detail throughout The General Theory, one aspect of this problem that he did not examine in detail is the vital role played by the flow of loanable funds in the economic system. Even though the flow of loanable funds does not determine the rate of interest and has very little to do with saving or investment, as Keynes clearly understood, [Blackford (2016b)] he did not seem to understand that this does not mean the flow of loanable funds can or should be ignored. Just as the flow of investment changes the stock of capital in the economy over time, the flow of loanable funds changes the stock of debt in the economy over time. And even though small changes in neither the stock of capital nor the stock of debt seem to affect the system in a significant way in the short-run, both can be expected to have dramatic effects on the system in the long run.

There is nowhere to be found in Keynes an analysis of the relationship between the flow of loanable funds and the accumulation of debt comparable to his analysis of the relationship between the flow of investment and the accumulation of capital. Nor, as far as I know, is such an analysis to be found anywhere in the writings of mainstream economists (Fisher and Minsky, yes; among mainstream economists, I am not aware of any) in spite of the fact that:

1. The creation of debt plays an essential role in achieving full employment by providing a mechanism through which purchasing power can be transferred from those who have it and are unwilling to spend to those who do not have it and are willing to borrow in order to spend. [Blackford (2014, Ch 3)]

2. At the same time, the most serious depressions involve financial crises that have at their root the inability to service debt.

3. It is obvious, or at least it seems obvious to me, that if the institutions of society are such that an increasing debt relative to income is required to achieve full employment in the short run, eventually debt service must overwhelm the system and cause a financial crisis that will make it impossible to sustain full employment in the long run.

It also seems obvious to me that herein lies the cause of major financial crises and depressions, and, yet, not only do mainstream economists ignore the dynamics of Keynes’ long-period problem of saving, they pay no attention to the circumstances in which full employment can be achieved only through an increase in debt relative to income in spite of the fact that this is exactly the kind of situation that existed leading up to the Great Depression and to the economic stagnation we see today. [Blackford (2014, Ch 3 and Ch 4; 2016b)]

Summary and Conclusion

The failure of mainstream economists to understand Keynes’ analysis of the long-period problem of saving—combined with their failure to understand the role played by increasing debt relative to income in creating financial crises—has resulted in the adoption of economic policies over the past forty years that have inhibited consumption and promoted saving and debt to such an extent that we can no longer achieve full employment in the absence of an increase in debt relative to income. This is a situation that, in the long run, inevitably leads to the kinds of financial crises we saw in 1929 and 2008—crises that lead to the kinds of economic, social, and political upheavals we faced in the 1930s, the last time we saw the kind of economic stagnation we see developing throughout the world today.

I find this particularly disturbing in that few economists seem to realize that the US economy did not recover from the Great Depression of the 1930s. What actually happened was the New Deal came along and the government took over the economic system during the war. There were 8.1 million unemployed in 1940 and the unemployment rate did not fall below 14% until 1941. It wasn't until 1943 that this rate fell below its 1929 level and by then the military had increased by over 8.5 million. In other words, the Great Depression did not end until wage and price controls were in place, consumer goods were rationed, the production of consumer durables ceased, the top tax rate was increased to over 90%, federal expenditures rose to 44% of GDP, and we increased the size of the military by more than the number of unemployed in 1940.

It seems quite clear to me that it was the institutional changes that occurred during the New Deal and WW II that took us out of the Depression and led to the prosperity that followed the war, not an economic ‘recovery’ as such. The system that emerged from the war was not the laissez faire system that led us into the Depression, and it was the institutional changes that ended laissez faire and allowed the income share of the top 10% to fall from 43% to 33% of total income, the top 1% from 16% to 11%, and non-federal debt to fall from 140% to 67% of GDP that led to the economic prosperity that followed the war—a prosperity that lasted until we began to undo those institutional changes in the 1970s. [Blackford (2014, Ch 3 and Ch 4; 2016b)]

When I look at the circumstances leading up to the financial crisis that began in 1929 and the depression that followed I find the only fundamental differences between then and now are the existence of social insurance programs, financial regulation, the size of government relative to the size of the economy, and the fact that policy makers had (for the most part) the good sense in 2007-09 not to follow the disastrous policies followed in 1929-33. Today we see the powers that be in the developed economies attempting to roll back the social insurance programs that have saved us so far from the kinds of deprivations experienced during the Great Depression, and we also see the beginnings of the same kinds of political and social upheavals we saw in the 1930s. [Blackford (2014, Ch 3 and Ch 4; 2016b)] This does not bode well for the future.

Institutional change as a result of depression and war is hardly an optimal way to resolve Keynes’ long-period problem of saving. It makes much more sense to affect those changes directly, but this cannot be accomplished in the absence of a clear understanding by mainstream economists of what the problem is. There can be little hope for the future until mainstream economists are able to look beyond their mathematical models, put aside their ideological blinders, and look at the long-period problem of saving in a way that leads to an overwhelming consensus within the discipline of economics to the effect that Keynes’ conclusion with regard to this problem—that consumption is the driving force for economic growth, not saving—is right and the failure of mainstream economists, both Keynesians and anti-Keynesians alike, to address this problem directly is a serious mistake. Only then will it be possible to affect the institutional changes needed to solve the long-period problem of saving directly, hopefully, in a way that will help to avoid yet another world-wide conflagration that, in this nuclear age, is likely to be even more horrific than the one that began on September 18, 1931 and reached its climax on August 6th and 9th, 1945.

References

Blackford, G. H., 2014, Where Did All The Money Go? How Lower Taxes, Less Government, and Deregulation Redistribute Income and Create Economic Instability, Amazon (Ch 1, Ch 2, Ch 3, Ch 4, and Ch 12 at Real-World Economics).

, 2016a, “A Note on Keynes’ General Theory of Employment, Interest, Money, and Prices,” Real-World Economics.

, 2016b, “Liquidity-Preference/Loanable-Funds and The Long-Period Problem of Saving,” Real-World Economics.

Fisher, I., 1932, Booms and Depressions: Some First Principles, New York: Adelphi Company. (PDF)

Hansen, A. H., 1953, A Guide to Keynes, New York: McGraw Hill.

Hicks, J. R., 1937, “Mr. Keynes and the ‘Classics’; A Suggested Interpretation,” Econometrica, pp. 147-159.

Keynes, J. M., 1936, The General Theory of Employment, Interest, and Money, Rendered into HTML, 2003, by Steve Thomas, University of Adelaide Library.

Marshall, A., 1961, Principles of Economics, 9th edition, New York: Macmillan. (PDF)

Minsky, H., 1986, Stabilizing an Unstable Economy, New York: McGraw-Hill.

Patinkin, D., 1956, Money, Interest and Prices: An integration of monetary and value theory, Evanston, IL: Row, Peterson and Company.

Robertson, D. H., 1936, “Some Notes on Mr. Keynes' General Theory of Employment,” Quarterly Journal of Economics, 51, 168-91.

Samuelson, P. A., 1947, Foundations of Economic Analysis, Harvard University Press.

Endnote

* This note provides a synopsis of the arguments and analysis with regard to the Crash of 2008 and the economic stagnation we see today that are presented in my eBook, Where Did All The Money Go?, and the two papers listed in the References.

Real-World Economics

Income, Debt, and Keynes’ Long-period Problem of Saving*