Scott Sumner  

Ralph Hawtrey

Against Recent Events... The Sheepskin Nightmare...

When I was appointed director of the monetary policy program at Mercatus, I was given the opportunity to name the "chair" I will occupy. Tyler Cowen made several suggestions, from which I chose Ralph Hawtrey, an outstanding British monetary economist from the interwar period. Hawtrey is not necessarily my favorite monetary economist (I also like Hume, Friedman, Fisher, McCallum, Hall and many others.) But I think he nicely fits a market monetarist oriented program. Other programs have been named after Milton Friedman, and if I had chosen Friedman it would have been harder to distinguish market monetarism from old-fashioned Friedmanite monetarism.

For those not familiar with Hawtrey, here are a few quotations. The first is from Currency and Credit, published in 1919, but this quote is taken from the longer 1930 edition (p. 451-52.) It comes right after Hawtrey discusses the fact that the value of gold (in terms of purchasing power) varies under a gold standard, and also discusses how index numbers can help guide monetary policy:

But then what allowance is to be made for a general scarcity or a general abundance of commodities? If the consumers' outlay be constant, the index number will be raised by scarcity and depressed by abundance. If the index number be constant, the consumers' outlay will be raised by abundance and depressed by scarcity.

The better alternative seems to be to aim at making the consumers' outlay constant. But, of course, it must not be absolutely constant; it must vary with the population, and must also vary in some way with the quality of the work they do. If that ideal could be attained, the value of the monetary unit in terms of human effort would be kept fixed.

But this is not the only possible solution. The monetary unit is employed for the measurement of debts. . . .

Hawtrey then has a discussion of how stabilizing an index of prices would assure that creditors receive back the purchasing power they anticipated when making a loan. Then he continues:

Either view seems to be perfectly tenable, and neither seems necessarily to represent the last word of justice. Justice in this sense means no more than a code of rules dictated by expediency, and rival solutions are to be tested solely by their practical consequences.

When it comes to practical consequences, all the debtor and creditor ask is that they may know how they stand, that they may be secured against arbitrary or incalculable variations in the value of the monetary unit. If it approximates to a fixed amount of human effort, or to a fixed amount of wealth, or to something intermediate between the two, this is in all probability sufficiently accomplished.

The danger is that the unit may wander far beyond these limits. . . .

There is some ambiguity here. The term 'GDP' did not exist in 1919, and in context "consumer outlay" seems to connote something like GDP. (Earlier he talks of the need to exclude intermediate goods to avoid double counting, which suggests he was thinking in terms of final goods.) I can't make sense of the phrase "quality of the work they do." Overall, here's how I read this:

1. Hawtrey considers lots of related policy rules, including price level, NGDP/person and wage targeting to be defensible, and all to be far superior to a regime where dramatic changes in the value of money occur.

2. Hawtrey is a pragmatist, who believes that policy rules should be adopted on practical (perhaps utilitarian?) grounds.

3. Toward the end he seems to lean toward something like NGDP targeting, wage targeting, or Selgin's productivity norm as being preferable to price level targeting.

Wage targeting can be justified in models where nominal shocks combined with sticky wages cause business cycles. Hawtrey seems to have had a sticky wage view of the cycle. Here's a quote from The Gold Standard in Theory and Practice:

Real monetary equilibrium in any single country requires the price level to be in harmony with the wage level, so that the margin of profit is sufficient, but not more than sufficient, to induce full activity and full employment. (Hawtrey, 1947, p. 45.)
When wages are out of equilibrium they tend to rise or fall (or rise at faster or slower rates). Thus steady wage growth suggests monetary equilibrium. Hawtrey believed that central banks were responsible for producing monetary stability. From the same book:
The collapse of demand is another name for the appreciation of gold. It means the offer of less gold in exchange for commodities. And we may regard the responsibility of the Central Banks as arising from their control over the market for gold. (Hawtrey, 1947, p. 140)
In modern lingo, AD shocks are another name for monetary shocks. Old monetarists tended to focus on the supply of money, whereas Hawtrey focused on both the supply and demand for the medium of account (which was gold when he was alive.) Central banks could impact the demand for gold, and this indirectly impacted the broader money supply and price level. Today he would focus on the fact that central banks control the supply of the medium of account (base money), but would not favor targeting the monetary base.

In a recent post, David Glasner showed that Hawtrey's "Treasury View" has been widely misunderstood by Keynesians, and that Hawtrey had a fairly sophisticated view of monetary offset of fiscal stimulus:

What has been shown is that expenditure on public works, if accompanied by a creation of credit, will give employment. But then the same reasoning shows that a creation of credit unaccompanied by any expenditure on public works would be equally effective in giving employment.

The public works are merely a piece of ritual, convenient to people who want to be able to say that they are doing something, but otherwise irrelevant. To stimulate an expansion of credit is usually only too easy. To resort for the purpose to the construction of expensive public works is to burn down the house for the sake of the roast pig.

That applies to the case where the works are financed by credit creation. In the practical application of the policy, however, this part of the programme is omitted. The works are started by the Government at the very moment when the central bank is doing all it can to prevent credit from expanding. The Chinaman burns down his house in emulation of his neighbour's meal of roast pork, but omits the pig.

I would encourage people to read Glasner's excellent analysis of this paper; here is how David concludes:

From the standpoint of pure monetary analysis, notwithstanding all the bad press that the "Treasury View" has received, there is very little on which to fault the paper that gave birth to the "Treasury View."
Back in Currency and Credit, I found this gem (p. 448):
Primitive man clung to his religious rites for fear he should neglect the progress of the seasons and the annual round of agricultural operations. The currency experts of the nineteenth century, with similar prudence, sanctified metallic money with an almost religious taboo.
Does that sound like something Keynes said?

Comments and Sharing

CATEGORIES: Monetary Policy

COMMENTS (11 to date)
Brian Donohue writes:

Great choice, great post.

Andrew_FL writes:

Well, without knowing all your options I can't say for sure if I would have made the same choice (but then, if it was anyone another than you, it would be a different research program in the first place) but Hawtrey seems like a good choice to me.

I do note with apprehension here, however, Hawtrey's use of "credit creation."

The distinction between created credit and transfer credit is an important one.

Even with purely transfer credit, the treasury view should hold. Savings that would have been supplied to finance the Government's borrowing would instead finance private Investment, or vice versa. This essentially corresponds to a constant MV rule, in which case changes in I and D (deficit) offset automatically (just as a consequence of that constraint and the accounting identity) and changes in T (the taxes removed from Y) Either result in corresponding changes in C or S (in the latter case resulting in changes in I) which also offset, again as a consequence of our constraint and the accounting identity. But no credit had to be "created." In fact our constraint forbid it.

Whether the treasury view holds in the short run turns out to be a question of how well financial intermediation works. Which is a consequence of the institutional framework, and the monetary policy pursued within that framework. Of course, it always holds in the long run.

Thomas Sewell writes:

I read "vary in some way with the quality of the work they do" related to keeping outlays constant as referring to taking into account changes in the usefulness of goods.

For example, if a pair of boots that previously lasted a year is later of a higher quality and lasts 14 months instead, then it's reasonable to expect the seller will charge more and the buyer will pay more, but less often. You may need to adjust your price comparisons over time to understand that prices will adjust with the quality of the work being done.

This is of course one of the traditional criticisms of the CPI, for example, that the CPI isn't adjusted to match quality of goods changes in a timely nor very accurate manner.

David Glasner writes:

Scott, I was very glad to hear that you chose to occupy the Hawtrey chair. I'm sure you will do great things in that capacity. And I'm also glad that you found my discussion of Hawtrey and the Treasury view helpful.

Michael writes:

Quality of work means productivity.

Jim Caton writes:

I'mm thrilled that you chose that name Scott. Truly in good spirit. Now if only we can get a Gustav Cassel chair :)

Ray Lopez writes:

It would be interesting to see Sumner compare and contrast the target NGDP framework to Selgin's Productivity Norm framework.

Andrew_FL writes:

@Ray Lopez-This may be of interest.

Scott Sumner writes:

Thanks everyone.

Thomas, I see how that could affect the CPI, but I'm not sure how it impacts NGDP.

Michael, So then what does he propose targeting?

Ray, George Selgin can answer that better than I can, but as far as I know the main difference between what I propose (targeting NGDP/(working age pop)) and what George proposes is that he also adjusts the target for for changes in other factors like capital, which I ignore.

Andrew_FL writes:

I actually think the Labor Productivity Norm is better than the Total Factor Productivity Norm, if only because there seem to be some measurement problems with TFP.

That, and for stable population it should converge to constant MV.

Scott Sumner writes:

Andrew, That sounds right to me, but I haven't studied the issue as much as George Selgin has.

Comments for this entry have been closed
Return to top