‘Savings’ can Increase Real Wages: A Reply to @asymptosis

I greatly admire Steve Roth’s (@asymptosis) blog as one of the best on theoretical monetary issues. His post on the characteristics of money is pretty definitive.

I take issue with a post of his No: Saving does not Increase Savings not because its or its underlying themes are wrong bit because it is only part right.

if you disagree with this post’s headline, you are thinking (perhaps unconsciously) in the “Loanable Funds” model. And the loanable funds model is complete, incoherent …

Think this through with me:

Your employer transfers $100K from their bank account to yours to pay you for your work. You’ve saved.

But is there more savings in the banks? More money to lend? Obviously not.

Thinking in this double entry way in terms of net drains and accretions to the banking system as a whole is a very useful way of avoiding accounting error. Steve pits the issue in a clear way though this perspective is not new being that held by the earliest endogenous money banking theorists (the earliest reference I can find is James Pennington in 1826, and it was repeated in terms nearly identical to Steve by Bostedo in his famous review of Bohm-Bawerk in 1901)..

So what is the problem? Well thinking solely in these terms makes the economy seem like a zero sum game where only the transfers of money matters. It is the same error I think made by Basil Moore and more recently Robert Barro in denying the Kenysian spending multiplier; in that it somehow manages to conjure ‘something out of nothing.’ Yet the capitalist process does manage to conjure more outputs than inputs, it creates a surplus and a profit from that surplus. If those profits are reinvested in capital formation rather than consumed we have economic growth.

The ‘monetary only’ perspective is in danger of ignoring five key issues:

  1. Value – it neglects value formation through production;
  2. Assets – it neglects the distinction between value formation through production and asset price speculation;
  3. Velocity – it neglects the importance of the velocity of money on prices and on the calculation of return on capital advanced. The higher the velocity of money the lower nominal profits needed to achieve a rate of return on capital advanced, even if the stock of money is unchanged.
  4. Profits – it neglects the importance of the link between profits and investment.
  5. Capital – it neglects the need for capital formation from savings in order to expand value formation.

An economy where 100% of profits were consumed rather than reinvested would see a boom in effectual demand that would be translated into inflation as the pace of production would not keep up with the demand for goods. None of this is incompatible with an endogenous money perspective. You may claim that future demand would create the demand and savings ex-poste through the S=I identity. However that rather dodges the point on how the loans would be financed? All loans must ex ante be financed through either enhanced equity of the lender (savings) or loans from another party (at extra costs) – eventually after several rounds of lending through lending the cost becomes prohibitive and all lending must be financed (ex ante) through savings. It is often poorly understood that the S=I identity is not an accounting identity that holds true at all times (Moore’s mistake) but in the medium run once monetary disequilibrium effects (multipliers) have run through. To quite from the General Theory.

‘An increment of investment in terms of wage-units cannot occur unless the public are prepared to increase their savings in terms of wage-units. Ordinarily speaking, the public will not do this unless their aggregate income in terms of wage-units is increasing. Thus their effort to consume a part of their increased incomes will stimulate output until the new level […] of incomes provides a margin of saving sufficient to correspond to the increased investment. The multiplier tells us by how much their employment has to be increased to yield an increase in real income sufficient to induce them to do the necessary extra saving …’ (Keynes 1936: 117)

Kaldor proved this though his ideas seem increasingly forgotten these days, he is a key player in this story. He had three key ideas, a model of growth (without production functions) where savings lead to capital formation, a nonlinear model of the business cycle where investment adjusts over time to the level of savings, and finally a proof of the Keynesian multiplier where through process analysis spending increased until the S=I identity was fulfilled and then the multiplier would cease.

This can be illustrated below:

A fully comprehensive, and stock flow consistent, treatment of a monetary theory of production must be able to rigorously account for the different ‘states’ of money flow in the circuit. This post is long enough so in the second part, and taking inspiration from Kaldor, I will attempt to classify these different ‘states’ of money in circulation and, in a step forward from his assumption of constant money prices, the impact of these ‘states’ in demand, capital formation, and prices. By adopting a comprehensive accounting framework there will be no ‘holes’ and we can derive accounting identities that must be true at all times whether at equilibrium or not that provide a proper bridge between pricing equations and the macroeconomy. From the identities describing any but one of the states of money you can derive the remaining one ‘ceritus paribus’.

My recent work looking at the ‘core contested ideas’ in economics has been looking in some detail at the history of debates on savings, the ideas of key players such as Turgot, Keynes, Kalecki, Bohm-Bawerk etc. It has shown be how very often theories talk past each other and never fully get to the bottom of an issue. There is an aspect of the ‘classical’ view of savings, which the Austrians, still hold to, which is worth ‘saving’ and of treated carefully is compatible with accounting identities and endogenous money.

Let me give a taste of the classification. I have avoided terms such as ‘liquidity preference’ as they can be ambiguous and the demand from money is in reality an amalgam of demand for different states of money some of which are demands for spending and some are not and which cannot wholly be encompassed into a single preference.

Class Monetary State Impact on Demand Impact on Investment
S1 Saving to Increase the Stock of Money (liquidity) without a plan to spend Decreases Demand Decreases Profits, Decreases investment in short terms. Creates a stock of money usable for investment in the longer term.
S2 Saving to defer consumption spending Decreases short term demand, though if used to save for a deposit on a loan the loan will increase demand once issued. Repayments of the loan will then decrease demand over term. Decreased demand for cheaper products will decrease investment in those. Increased investment in luxury goods.
S3 Saving to defer and increase consumption through capital formation As S2 Decreased demand reduces investment in the short term though lending will increase investment more than offsetting this.
S4 Saving to defer and increase consumption through asset purchases Decreases short term demand, though if used to save for a deposit on a loan the loan will increase demand once issued. Repayments of the loan will then decrease demand over term. As the asset supply is broadly fixed the impact of the spending will be to increase asset prices. For investment read speculation. Decreased short term demand more than offset by increased lending. So long as the asset price increases by more than the return on capital will fuel a bubble until the asset becomes unaffordable.

Only one of these can increase real wages. S3 – through capital formation, expansion of the market and introduction of new techniques. S1, S2 and S4 all can in different ways lead to collapse on effectual demand and capital destruction. Capital formation is just a transfer of bank reserves, but it increases value formation through putting idle account to work employing people. The issue is slightly complicated as financial intermediaries can pull off the Warren Buffet trick of ‘liquidity transformation’ ensuring the demanded outflow (S2+s3+s4) from a depositor is maintained whilst using deposits in excess of reserve requirements as S3 or S4 investments. This will be of little use for investment if there is a general increase in S1 savings as these intermediaries will be seeking safe assets rather than to make investments due to the collapse in demand.

The total stock of money in any portfolio must comprises of one of these four monetary classes. Therefore one can derive any of the four stock levels from the other three.

Hence we can turn to Bohm-Bawerks defence of the classical Tugot-Smith theory of saving/capital formation following an identical attack to Steve’s from Bostedo,

Mr. Bostedo .. says with special emphasis that every saving is only a transfer of purchasing power from the savers to other members of the community.

The fault in the .. It is that one of the premises, the one which asserts that a curtailment of “consumption for immediate enjoyment” must involve also a curtailment of production, is erroneous. The truth is that a curtailment of consumption involves, not a curtailment of production generally, but only, through the action of the law of supply and demand, a curtailment in certain branches. If in consequence of saving, a smaller quantity of costly food, wine and lace is bought and consumed, less of these things will subsequently and I wish to emphasize this word be produced. There will not, however, be a smaller production of goods generally, because the lessened output of goods ready for immediate consumption may and will be offset by an increased production of “intermediate” or capital goods.

The argument is not fully rigorous in that, unlike the Bostedo argument that inspired Keynes he does not allow for hoarding, nor does he distinguish between capital formation and speculation on assets. Subject to these modifications, and major modifications they are, the classical theory of capital formation is compatible with keysian theory and endogenous money. Saving matters.

Further Reading:

Anne Robert Jacques Turgot: The Importance of Capital Antoin E. Murphyin The Genesis of Macroeconomics December 2009 OUP

The Function of Saving Eugen Von Bohm-Bawerk Annals of the American Academy, Volume 17 (1901) 

Economic Writings of James Pennington 1826-1840 Psychology Press, 1997 –  114 pages

The Keynesian Multiplier Liquidity Preference And Endogenous Money Dr Paul Dalziel March 1995

Kaldor. N. “Speculation and Economic Stability.” Revi(‘w of /:’collolllic Studies, 1939, 7( I).

Kaldor, N. and Trevithick. J. “A Keynesian Perspective on Money.” Lloyd’s Bank Neviell’. Spring 1981,139.
Keynes, J.M. Gneral Theory of Employment, Interest and Money . London: Macmillan, 1936.


Moore, BJ, ‘The Demise of the Keynesian Multiplier: A Reply to Cottrell.” Journal of Post Keynsian Economics, Fall 1994, 17( I).