Steve Keen on Loanable Funds and Endogenous Money

Steve Keen gives a lecture below on the macroeconomics of loanable funds and endogenous money using his program Minsky.

Daniel Kuehn on Loanable Funds

Daniel Kuehn has a nice response to me on loanable funds:

Daniel Kuehn, “Keynes and Loanable Funds,” Facts and Other Stubborn Things, August 5, 2013.

First, I am happy to fold on a number of issues where I was wrong.

It is completely right to say that modern Keynesian economics should not be about who is more “faithful” to Keynes. Keynes made mistakes. He was not always right, and a proper Keynesian economic science must move beyond Keynes. I agree.

I was also wrong to imply that Daniel just self-identifies as a strict New Keynesian. Sorry about that!

Nor, when I cited an article in a previous post that happened to be called “Bastard Keynesianism,” did I mean to insult New Keynesians. On reflection, the term “Bastard Keynesianism” that was coined by Joan Robinson in 1962 to refer to the neoclassical synthesis (in addition to being rude!) gives the unfortunate impression that Keynesian economics is just about blindly following Keynes, which it certainly should not be.

And, yet, when it comes to the other issues I fear we may be talking past one another. Obviously, there is a flow of money into banks that represents funds people want to save, and in return they get an asset: either (1) the credit money we call demand deposits (or checking accounts or saving accounts) or (2) financial assets called time deposits.

But surely classical loanable funds theory is, fundamentally, a theory of interest rates, saving and investment. It makes many more claims than the simple observation that there is an annual flow and stock of savings.

Now I am sure Daniel is perfectly familiar with Keynes’s critique of loanable funds.

So the remarks that follow are really more for my benefit and other readers of this post.

Keynes’s critique of the loanable funds theory is summed up by Bill Mitchell:

“… the Classical belief [sc. was] that the household decision to save was determined by the preferences for current and future consumption mediated by the interest rate (the price that consumers traded current consumption for future consumption). Instead, … [sc. Keynes] considered aggregate saving was a positive function of national income.

So when national output and income rises, aggregate saving will rise. The amount of extra saving per dollar of additional disposable income is called the Marginal Propensity to Save (MPC). If the MPC = 0.20, then households will save 20 cents of every extra dollar of disposable income they receive.

The interest rate might have some influence on saving but Keynes considered the influence of changes in national income to the dominant factor determining the aggregate level of savings in any period.

The other consideration is that investment spending is a component of aggregate demand, which in turn, drives total national income in each period.

Taken together, these insights undermines the concept of a loanable funds market in the way conceived by the Classical economists. There could not be independent saving and investment functions brought together by movements in the interest rate as required by the loanable funds doctrine because investment drove income which influenced saving.

In Chapter 14 … of his General Theory of Employment, Interest and Money, he produced a diagram to illustrate his contention that this interdependency meant the loanable funds doctrine was a ‘nonsense theory.’”
Bill Mitchell, “Keynes and the Classics Part 6,” Billy Blog, January 24, 2013.

There is also the question of what information, if anything reliable, is communicated to businesses through interest rates about time preference.

And then we have more complicated issues about money and banking, such as endogenous money theory, relevant to the classical loanable funds.

Money saved adds to a bank’s reserves. But even at this point the standard story is flawed. Bank lending is not constrained in the way the standard theory requires. Loans create deposits (or new money), and most of the broad money stock is bank money held in the form of demand deposits. Prior monetary saving is not strictly necessary for investment. Then there is the issue of the mythical money multiplier.

At this point, however, we are simply revisiting some of the issues of the Krugman versus Keen debate on endogenous money from about a year ago, which I discussed here:

“Keen versus Krugman: The Great Debate!,” April 4, 2012.

Keen also gives a talk below that addresses some of these points.





Further Reading
“Keynes and the Classics Part 6,” January 24, 2013.

“Scott Fullwiler: Krugman’s Flashing Neon Sign,” Naked Capitalism, April 2, 2012.

Mises’s “Originary Interest”: Another Useless Real Theory of the Interest Rate

Mises’s definition is as follows:

“Originary interest is the ratio of the value assigned to want-satisfaction in the immediate future and the value assigned to want-satisfaction in remote periods of the future. It manifests itself in the market economy in the discount of future goods as against present goods. It is a ratio of commodity prices, not a price in itself. There prevails a tendency toward the equalization of this ratio for all commodities. In the imaginary construction of the evenly rotating economy the rate of originary interest is the same for all commodities” (Mises 2008: 523).

First, this is quite clearly a real theory of the interest rate, in which present real goods exchange for future real goods, or in other words where loans are imagined as occurring in natura. The “originary interest” is a ratio. Mises describes it as the “discount of future goods as against present goods” (Mises 2008: 521).

In terms of capital goods, the “originary interest rate” arising in Mises’s equilibrium world called the “evenly rotating economy” (ERE) would be the same as the Wicksellian natural rate of interest.

Mises asserts that there is “a tendency toward the equalization of this ratio for all commodities,” but this is unconvincing. Nor can there ever be a single “originary interest rate” outside of purely imaginary equilibrium states.

Mises has his own monetary or market rate of interest on loans called the “gross money rate of interest” (Mises 2008: 534). Mises conceives the “gross money rate of interest” as being determined by other factors in addition to originary interest, as follows:

(1) the entrepreneurial component: interest determined by the speculative element in money lending and the dangers involved (Mises 2008: 536–538).

(2) the price premium: an additional element in interest to take account of expected changes in inflation or the general price level (Mises 2008: 538–542).

In Austrian theory, unfettered market interest rates – or ideal laissez faire gross money rates of interest – are supposed to gravitate towards an “originary interest rate” or Wicksellian natural rate of interest, and thereby allocate resources effectively in an intertemporal sense. By this process, a monetary interest rate is supposed to move towards a natural rate of interest so that his coordinates resources and provides intertemporal coordination of investment projects with real resources.

But the natural rate of interest can only be a single rate inside general equilibrium (or in some other equilibrium state such as Mises’s “final state of rest” or the ERE). Outside of general equilibrium, there can be as many natural rates as there are capital goods commodities lent out. No monetary system where capital goods investments are made by means of money can hit the right multiple natural interest rates either on each in natura loan of various capital goods, because even though the banks’ monetary interest rates – even in a free banking system – might converge in a spread, there could be vast differences between the spread of banks rates and many individual commodity natural rates.

The Austrian theory of interest rates – either the monetary rate or the (alleged) real rate – is grossly unrealistic and flawed.

First, in any advanced capitalist economy, people are generally lending and borrowing money, not real goods. How can a convergence to a “real” originary rate on goods emerge when borrowing is not in barter loans, but in terms of money?

Secondly, the nature of monetary interest rates in a market economy, and even hypothetical free market ones, is not described by time preference theory.

In any capitalist economy, there will generally be a stock of money used to buy and sell assets on secondary markets (whether real or financial markets). This money can be diverted to use in lending or clearing of loans for capital goods investments. Even if one were to use loanable funds model, a decrease in liquidity preference can increase funds available for lending without a corresponding decrease in consumption, since it might be merely money previously used on secondary asset markets or dishoarded. Such shifts are merely just changes in the liquidity of assets held in a person’s portfolio, not changes in time preference. Therefore changes in monetary interest rates even in some hypothetical free market system need not necessarily communicate any significant information about time preference or even any meaningful information at all.

Even when people abstain from consumption and save money, it does not follow that their saving now will entail a consumption expenditure in the future. Resources may not have been consumed now, but it does not follow that the eventual output of those resources in a capital goods project will be demanded in the future.

To sum up, one can say that:

(1) a decision to save money now does not entail a future consumption purchase;

(2) saved money now need not be invested in capital goods projects, and may be used to buy secondary assets (either real or financial). A large stock of money is at any one time tied up in purchases and sales of secondary assets;

(3) money made available for capital goods investments may have simply been shifted from purchasing of secondary assets (either real or financial) and not from abstention from consumption, and there need be no change in time preference, only liquidity preference.

(4) changes in monetary interest rates, even in hypothetical free market economies, need signal no reliable information and indeed no information at all about time preference or real resource availability.

BIBLIOGRAPHY
Maclachlan, Fiona C. 1993. Keynes’ General Theory of Interest: A Reconsideration. Routledge, London.

Mises, L. 2008. Human Action: A Treatise on Economics. The Scholar’s Edition. Ludwig von Mises Institute, Auburn, Ala.

Greg Hill on “The Moral Economy: Keynes’s Critique of Capitalist Justice”

Hill (1996) is an important article on the fundamental economic issues in Keyes’s General Theory, and it set off an epic debate with the Austrian economist Steve Horwitz that can be read in Horwitz (1996), Hill (1996a), Horwitz (1998), and Hill (1998).

I will concentrate below on Keynes’s view of saving and loanable funds theory.

Hill (1996: 34) points out that, while Keynes certainly was a supporter of capitalism in the general sense of allocating scarce factor inputs to provide consumer goods, he argued that capitalist economies do not necessarily provide a high enough level of private investment and employment, nor that all income distribution generated by capitalism is just.

Thus Keynes’s argument for managed capitalism is both an economic and a moral case.

At the heart of Walrasian neoclassical economics is the price system. Walras’s metaphor for modeling a capitalist economy is that of the auctioneer who can announce quantities and prices of goods, and adjust prices in the process of tâtonnement until a set of market clearing prices is discovered to achieve equilibrium (Hill 1996: 35). The metaphor is utterly unrealistic, since the auctioneer must have perfect or near perfect information about different plans and trades, all expectations are fulfilled, the process is almost timeless, and stripped of the fundamental uncertainty which characterises economic life.

Under such a theoretical framework, neoclassical economics envisages a just earning of all wages, profits, and interest from all work, investment, and thrift respectively (Hill 1996: 36).

Keynes strongly challenged this paradigm.

First, the link between saving and investment. It is assumed in the neoclassical theory that saving will induce capital goods investment (Hill 1996: 38). Real resources “saved” in the sense of not being used to make consumption goods consumed today will be used in capital goods projects to make goods in the future.

But the link is not automatic nor reliable. Hill quotes a classic a passage on the general theory on the reality of any act of saving:

“An act of individual savings means –so to speak – a decision not to have dinner to-day. But it does not necessitate a decision to have dinner or to buy a pair of boots a week hence or a year hence or to consume any specified thing at any specified date. Thus it depresses the business of preparing to-day's dinner without stimulating the business of making ready for some future act of consumption.” (Keynes 1936: 210).

There is no necessary reason why an act of saving of money today must entail a future act of consumption (Hill 1996: 40), either in nominal money terms or real purchasing power terms. Money income may be spent on (1) goods and services, (2) second hand assets (whether real or financial), or (3) hoarded.

Certainly as a person’s income rises, they are more likely to spend money on (2), rather than (1).

Furthermore, the inducement to investment is complex and a function of demand, expected demand and expectations. Saving may well result in reduced demand and expected demand, and in reduced capital expenditures.

Of course, loanable funds theory gives capitalist apologists an imagined escape hatch here, as Hill notes:

“Neoclassical and Austrian economists reject this deflationary interpretation of the act of saving. In their version of Keynes’s parable, when someone reduces his consumption spending (e.g., by no longer dining out), the total supply of savings rises, and the rate of interest falls. This reduction in the cost of borrowing increases the number of profitable investment projects, and, in response, entrepreneurs increase their level of investment spending. In short, the act of saving supplies the wherewithal necessary for investment, and the falling rate of interest, which signals the public’s desire to trade current for future consumption, assures a commensurate increase in investment.” (Hill 1996: 40).

Keynes’s own critique of this was, firstly, that an individual act of saving may in fact decrease the aggregate level of savings: one saving decision reduces the incomes and hence saving of a business, and then has knock-on effects as that business reduces its own spending and hence savings of other businesses (Hill 1996: 40).

This brings out the well known fallacy of composition called the paradox of thrift: if everyone increases their savings, then the aggregate effect may be in fact decreased income and ultimately saving and investment (Hill 1996: 40).

A secondary criticism is the alleged coordinating role of interest rates in loanable funds:

“suppose that there is a flow of saving per year, and a flow of investment per year, and that the rate of interest adjusts so as to bring these two flows into balance. Now, if the interest rate were only required to equilibrate these flows of new lending and borrowing, it might well be able to perform the coordinating role assigned to it by the neoclassical school. There is, however, another important dimension to the problem, for the market in which new bonds are issued is the same market in which existing bonds are traded. And the very same scheme of interest rates that must balance the supply and demand for new bonds must also balance the supply and demand for old bonds. What would happen, then, if there were a conflict between 1) the rate of interest that would balance the flows of new saving and investment and 2) the rate of interest that would balance the supply and demand for existing bonds? According to Keynes’s account, the outcome will be determined by decisions concerning the existing stock of bonds because, at any given moment in time, the quantity of old bonds that can be released onto the market dwarfs the quantity of new bonds entering the market, just as the stock of money being held in anticipation of a fall in bond prices dwarfs the quantity of new additions to savings. Against the massive, preexisting stocks of old bonds and of money poised to enter the market in response to a change in the interest rate, the relatively small flows of new lending and borrowing can have little effect. It is because the rate of interest must balance these great stocks of existing wealth that it cannot, at the same time, effectively coordinate the flow of saving and investment.” (Hill 1996: 41–42).

The crucial issue here is the way interest rates in a real world capitalist economy involve much more than just some equating of investment with saving. There is vast stock of money and a vast stock of secondary financial assets bought with money as well as lending for capital goods projects.

Once one adds to this the possibility that business expectations can be shattered, it follows that lowering interest rates will not necessarily induce sufficient investment to create high employment and growth. A better solution is increasing demand and expected demand by greater spending.

In short, Keynes “turned the virtue of thrift on its head” (Hill 1996: 43) and undermined the classical argument for inequality of wealth.


BIBLIOGRAPHY
Hill, Greg. 1996. “The Moral Economy: Keynes’s Critique of Capitalist Justice,” Critical Review 10: 411–434.

Hill, Greg. 1996a. “Capitalism, Coordination, and Keynes: Rejoinder to Horwitz,” Critical Review 10: 373–387.

Hill, Greg. 1998. “An ultra-Keynesian Strikes Back: Rejoinder to Horwitz,” Critical Review 12: 113–126.

Horwitz, S. 1996. “Keynes on Capitalism: Reply to Hill,” Critical Review 10.3: 353–372.

Horwitz, S. 1998. “Keynes and Capitalism One More Time: A Further Reply to Hill,” Critical Review 12: 95–111.

Keynes, J. M. 1964 [1936]. The General Theory of Employment, Interest, and Money. Harvest/HBJ Book, New York and London.

Greg Hill versus Steve Horwitz: A Keynesian–Austrian Debate

The Austrian Steve Horwitz and the Keynesian Greg Hill had a debate on the pages of Critical Review as follows:

Hill, Greg. 1996. “The Moral Economy: Keynes’s Critique of Capitalist Justice,” Critical Review 10: 411–434.

Horwitz, Steven. 1996. “Keynes on Capitalism: Reply to Hill,” Critical Review 10.3: 353–372.

Hill, Greg. 1996a. “Capitalism, Coordination, and Keynes: Rejoinder to Horwitz,” Critical Review 10: 373–387.

Horwitz, Steve. 1998. “Keynes and Capitalism One More Time: A Further Reply to Hill,” Critical Review 12: 95–111.

Hill, Greg. 1998. “An ultra-Keynesian Strikes Back: Rejoinder to Horwitz,” Critical Review 12: 113–126.

While I will not cover every detail of the debate, two important aspects of it were the issues of (1) the coordination of saving and investment, and (2) loanable funds theory.

Hill notes that, as Keynes argued, the decision not to spend one’s income today on goods or services does not entail that the money will be spent in the future on goods or services (Hill 1998: 114).

Horwitz appeals to a loanable funds model, in a rather idealised form, in his critique of Hill. Under the loanable funds theory, when people reduce consumption, the resulting savings add to loanable funds and this is supposed to lower interest rates and induce more capital investment in production of goods that will be available at some point in the future. The hidden assumption here is that an individual’s increased saving will necessarily increase total saving, for it is total saving that must be increased to reduce the rate of interest (Hill 1996a: 374). But even if we assume that the money saved by a potential consumer is made available for capital goods investments, the money not spent by the consumer reduces the income and savings of a business where he or she would have spend the money, and the income and savings of the business’s employees and suppliers. Therefore the addition to business savings and savings of those who earn income from the business will be prevented by loss of income from the first act of saving of the consumer, and the total amount of savings need not be higher (Hill 1998: 116). When such a process occurs throughout an economy, in the aggregate there need be no increase in saving, but reduction in demand deposits, and reduction in the broad money stock. Therefore the interest rate need not fall, and the supposed inducement of more investment will not happen. Thus an increase in current savings need produce “no inducement to expand future output in the absence of an order for future delivery” (Hill 1998: 115).

Austrians might claim that a lower interest rate – or a lower price of credit – will induce greater investment when interest rates fall, but this does not necessarily follow in a world where business faces uncertainty, where expectations are subjective, and where demand for investment credit can collapse or be stagnant.

And there is also a fundamental flaw in the whole loanable funds model: the fact that a great deal of what we call “saving” is spending of money on secondary real asset markets and, above all, on secondary financial asset markets. The decision not to spend money on consumption goods today does not mean the money is necessarily transferred to banks that finance capital goods investments. Often, especially in the case of the rich and very rich, the money is used to buy financial assets, and may be diverted to exchanges between buyers and sellers of such financial assets for significant periods of time. The purchase of a stock, bond, or financial instrument on a secondary financial asset market does not make that money available for capital goods investments per se; nor does it lower interest rates. In other words, there is a devastating flaw running through the whole loanable funds model: the assumption that money not spent is going to be simply put in a financial institution that lends the money for real investment.

The blogger MGM on the short-lived “Austrian Economics” blog expresses a crucial observation on this point:

“… savings find their way into the financial sector, because financial assets possess a great deal of liquidity. And depending on one’s appetite for risk, one can attempt to sacrifice a little liquidity for the possibility of capital gains (speculation); but because most financial assets have orderly markets, it is relatively easy to sell these assets for money. Capital goods, however, are not easily resalable (liquid).

For Post Keynesians, the financial sector is very different from the industrial sector. The financial sector deals principally with liquidity, and aims to provide people with liquidity (savers). The industrial sector, on the other hand, deals with real tangible (not easily substitutable) capital goods. These goods do not provide liquidity, because they cannot easily be sold. People who deal with capital goods must therefore look to its prospective yield and not its liquidity properties. These people are generally capitalists, and not savers.”
“The Horwitz and Hill Debate: Or, Why the Austrians are Wrong about Financial Markets,” Austrian Economics, March 27, 2011.

Moreover, the banking and monetary system is endogenous, which means that it generates new money in response to the demand for (1) credit or (2) demand deposit money. Hill is absolutely right to stress that modern banking systems create credit in excess of savings and prior monetary saving is not needed to back investment (Hill 1996a: 381). Assuming resources are available, new monetary saving is therefore not even necessary for increased credit creation and investment. In a world of vast international trade, industrial sectors with unused excess capacity, and idle resources, even at a high level of employment, capitalist systems can still provide elasticity of production of many goods without serious inflation.

But the demand for investment credit is still dependent on many things other than a crude supply and demand curve for money, such as expectations of future profit, the level of demand, sales volume, expectations of future sales and orders, and so on.

Horwitz also thinks that flexible wages and prices will prevent, or at least be the solution for, unemployment when saving exceeds investment. Here Hill notes that Horwitz misunderstands Keynesian theory, because Keynes in fact argued that, even if wages and prices were perfectly flexible, involuntary unemployment would exist (Hill 1996a: 377). The most devastating response Hill has to Horwitz’s flexible wages and prices model as the cure for unemployment is the disastrous debt deflation that results from wage and price reductions (Hill 1996a: 378).


Links
Robert Vienneau, “Steven Horwitz and Post Keynesians,” Thoughts on Economics, June 1, 2008.

“The Horwitz and Hill Debate: Or, Why the Austrians are Wrong about Financial Markets,” Austrian Economics, March 27, 2011.

Dan Kervick, “Do Banks Create Money from Thin Air?,” New Economic Perspectives, June 3, 2013.


BIBLIOGRAPHY
Hill, Greg. 1996. “The Moral Economy: Keynes’s Critique of Capitalist Justice,” Critical Review 10: 411–434.

Horwitz, Steven. 1996. “Keynes on Capitalism: Reply to Hill,” Critical Review 10.3: 353–372.

Hill, Greg. 1996a. “Capitalism, Coordination, and Keynes: Rejoinder to Horwitz,” Critical Review 10: 373–387.

Horwitz, Steve. 1998. “Keynes and Capitalism One More Time: A Further Reply to Hill,” Critical Review 12: 95–111.

Hill, Greg. 1998. “An ultra-Keynesian Strikes Back: Rejoinder to Horwitz,” Critical Review 12: 113–126.