It is sometimes said that capitalists consume a lot and the theory of underconsumption fails.

For example, Federal Reserve paper (h/t Winterspeak) Disentangling the Wealth Effect: A Cohort Analysis of Household Saving in the 1990s – although not suggesting a failure of underconsumption – says that in the ’90s, personal saving rate fell a lot due wealth effects from a rise in the market value of equities held by high-income earning households (although suggesting alternate interpretation as well) inducing them to consume more.

But some on internet discussion groups seem to have interpreted it as a failure of the underconsumption theory.

Before I say what I want to say, some preliminaries: In Keynesian theories, there are two things – propensity to consume and propensity to save. A lot of people erroneously think that these add to 1. I will go further and say that a theory which considers the interaction of stocks and flows with a notion of a propensity to save will end up with paradoxes. Rather saving – household saving in particular should be seen as a result of a model which traverses in historical time between one configuration of stocks and flows to another. With consumption depending on income and wealth, the notion of propensity to consume is more appropriate than the somewhat inferior “propensity” to save. Or one may say that the propensity to save is a derived concept.

What is the underconsumption theory? It can be conjectured in various ways. One is to say that the propensity to consume out of wages is higher than the propensity to consume out of interest income, dividends and capital gains. Another is to say that low income earners have a higher propensity to consume out of wages than high income earners. This is then given a policy angle and the underconsumptionists then claim (rightly in my opinion) that a better distribution of national income will lead to stronger effect on growth because low income earners will consume more and have multiplier effects on output – in contrast to “supply-side” policies.

Back to the Federal Reserve paper. Some seem to have interpreted it as saying that it is a blow to the theory of underconsumption citing low rate of saving for high income earners. There are several things wrong with this. The fact that high-income earners spent a lot in some periods doesn’t mean it is a better policy than aiming for a better distribution of income, especially when the former is more unpredictable because of the unpredictability of stock-markets, while the latter is more direct. Although this point is more important, the citing of low saving of high-income earners in some periods following high capital gains should be dismissed. Low saving was not the result of a low “propensity to save” – which is a derived concept but the result of higher consumption due to rise in wealth than due to a rise in “propensity to consume” by high-income earners.  

So if household consumption is given by

C = α₁·YD +  α₂·W_-1

where C is consumption, YD is disposable income, and W is wealth. α₁ and α₂ are the propensities to consume out of income and wealth, respectively. The subscript -1 indicates it is the value for the previous period. The wealth term also includes capital gains at the end of previous period.

Saving is given by

S = YD – C

Now, as per national accounts, income doesn’t include capital gains, although taxes on capital gains may reduce YD but let’s ignore that. This doesn’t mean capital gains are irrelevant, because consumption behaviour depends on capital gains. So the rate of saving  S/YD is given by

S/YD = (1 – α₁) – α₂·W_-1/YD

which may hit zero or even turn negative if there is a large rise in W_-1 because of capital gains even if the parameters α₁ and α₂ have not changed. So for example α₁ may be something like 0.5 and yet the rate of saving may be negative.

In other words, the data of the 90s doesn’t disprove the fact that low-income earners’ propensity to consume out of income is higher than high-income earners – it just shows that some high income earners’ saving may have gone negative because of higher consumption due to capital gains in some periods.

So what if there is a fair distribution of income? Since low-income earners have a higher propensity to consume, less inequality will lead to a higher output and national income than otherwise.

[This is further to my previous posts Central Bank LSAP: Is The Money Stock Supply-Determined Or Demand-Determined? and On Effects Of QE.]

Central bank asset purchases or LSAPs increase the stock of money aggregates compared to the counterfactual when there is no QE. Is there an excess of money?

These are two different things.

The notion of “excess money” is usually associated with Monetarism where the supposed “excess” is eliminated by rise in prices of goods and services. Excess money causes prices rise as much to reduce the value of money to reconcile whatever economic agents want to hold so that money demand = money supply where supply is given exogenously.

Which of course is quite wrong.

The reason it is wrong is that while – as argued in the previous posts – QE causes the stock of money to rise, there is no excess because prices and perhaps even quantities in financial markets (which is different from the market for goods and services) adjust so that the stock of money is still demand determined.

Of course that is not to say that the stock of money is the same as the counterfactual – i.e.,  the stock of money in a QE world is different from what it would have been in the absence of QE. So the stock of money is not exogenously given and adjusts to the demand for money as argued in the previous post and this demand depends on various things such as expected returns on imperfect substitutes – which QE influences.

In the markets for goods and services, there’s hardly an effect because prices are set by producers according to their costs.

Of course there needs to be a few qualifications – in the market for primary products, myths and speculation can lead to a rise in prices – such as the price of oil and this can have inflationary effects but the world we live in is not one in primary products and the original Monetarist idea highlighted at the start of this article is still misleading.

Of course none of this is implies that LSAP/QE has less effect on prices of financial markets – it can indeed cause some unwanted booms.

Another thing is that I mentioned in the previous blog post that supply of money influences demand but doesn’t determine it. Does that mean that the central bank has a “control” on the stock of the money within some limits? Well, not really. Control means influencing whatever is said to be under control to move in the desired direction and this is not always so. For example if the central bank were to behave like what the Bank of England was experimenting in the 70s under Monetarists’ influence, and sell bonds in large quantities in the financial markets, economic units – such as the non-banking sector may expect bond prices to fall and desire to increase their holding of money (as in deposits). This will lead them to sell the bonds to the banking sector which is prepared to buy the bonds acting as dealers in the bond markets. So the outcome may be opposite of the objective with which the central bank started with.

The blog Fictional Reserve Barking … Against Fictions And Other Tall Tales has a nice post titled: On the (ir)relevance of the money multiplier model: The Fed view.

In the post, the blogger Circuit quotes Robert Hetzel who says:

Starting in mid-December 2008 when the FOMC lowered its funds-rate target to near zero with payment of interest on bank reserves, the textbook reserves-money multiplier framework became relevant for the determination of the money stock. The reason is that the Fed’s instrument then became its asset portfolio, the left side of its balance sheet, which determined the monetary base, the right side of its balance sheet. As a result, from December 2008 onward, the nominal (dollar) money stock was determined independently of the demand for real money. Although the reserves-money multiplier increased because of the increased demand by banks for excess reserves, the Fed retained control of M2 growth. Even if banks hold onto increases in excess reserves, the money stock increases one-for-one with open market purchases. (2012:237) (emphasis added)

This post is inspired by the nice comment by JKH in the same post.

JKH says:

Excellent post.

I think Hetzel is wrong though.

What is at work in QE is better described as a money duplication process rather than a money multiplier process. It’s entirely different.

The correlation between base changes and M2 changes during QE has nothing to do with the reserve ratio calculation that is part of the money multiplier math.

It has to do with the fact that non-banks were the ultimate source for most of the assets acquired by the Fed under QE. To the degree that’s the case, there is a 1:1 duplication of reserve expansion and M2 expansion at the point of transaction origin. Non-bank bond sellers basically convert their bonds to M2 at source.

Subsequent commercial bank balance sheet changes may change the one-to-oneness that appeared at origination, but that also has nothing to do with money multiplier dynamics.

So in total this has nothing to do with a standard money multiplier argument. The two should not be confused.

I have a few more things to add on Hetzel.

When the Federal Reserve buys assets such as US Treasuries and agency debt and mortgage-backed securities from the private sector, it increases the stock of money aggregates such as M1 – as most of the ultimate sellers are non-banks, even though the Federal Reserve purchases the bonds via reverse auctions from primary dealers.

Now whether the money multiplier story works or not (it doesn’t!), it superficially looks as if the Federal Reserve is determining the stock of money – supporting Hetzel’s view – again superficially.

But does it?

In Post-Keynesian monetary theory, economists say that the stock of money is “demand-determined” and Hetzel’s arguments seems to be against this view. However what Hetzel forgets is that while the Federal Reserve influences the stock of money, it is still demand-determined. This is because money-demand depends on agents’ portfolio preferences. If agents have “excess” stock of money, they may reflux this by reducing their loans toward the banking system – just like Post-Keynesians claim. The Federal Reserve QE Education page New York Fed 101: The Federal Reserve’s $600 Billion Treasury Purchase Program (Called by some QE or QE2) itself says so:

In fact since QE leads to higher asset prices, economic units may in fact borrow more funds to invest (read: speculate) in the stock markets.

Also note that “money-demand” is dependent on various things such as expected returns on substitutes. To the first approximation we may say that the stock of money was determined by the Federal Reserve since it has an eye on the monetary aggregates (although not targeting in the Monetarist sense). But once the process is set in motion, it then becomes endogenous because of behaviour of economic units.

Only in the limited case of settlement balances can we say that a monetary aggregate (monetary base or MB in this case) was set by the central bank.

The correct way of seeing this is via using Wynne Godley’s asset allocation model. In this what the variable M_d is still decided by household behaviour – which depends on portfolio preferences, wealth, income and expected returns on all assets. LSAP works by reducing the supply of long-term bonds and hence reducing long-term yields via the “preferred-habitat” theory and hence increasing demand for other assets via imperfect asset substitution and changing their prices.

This itself has an influence on money-demand (because money-demand also depends on expectations of returns of other assets) but the amount of asset purchases by the Fed doesn’t determine the stock of money at any point in time.

Hence Hetzel is both wrong by appealing to the money multiplier mechanism and even if he hadn’t his other point about the stock of money being supply-determined is incorrect.

You can read Nick Edmonds’ blog on how this works – he has the most precise way of describing QE/LSAP.