This week has seen an almighty blog ding dong between prominent New Keynesians.
Williamson kicked it off here on the 27th Nov. Arguing is highly mathematical language that QE has been deflationary.
the effect of QE is to lower the liquidity premium (collateral constraints are relaxed) which…will lower inflation and increase the real interest rate.
An idea that has been floating around in post-keynsian circles for some time. See Cullen Roache and Francis Coppola for example., with particular stress on the role of QE in providing safe collateral. Narayana Kocherlakota had briefly made such a claim in 20011 before withdrawing it.
Horror erupted with arguably the leading new Keynesian monetary theorist being threatened with ejection from the club, as summed up by Noah Smith
In a testy response, Nick Rowe called Williamson’s post “horribly wrong,” lamenting: “What the hell has gone wrong with some of the best and brightest in economics?” Brad DeLong then jumped in, accusing Williamson of mistaking an unstable equilibrium for a stable one. Paul Krugman echoed that accusation.
There were further follow ups by Williamson here, calling it a ‘teachable moment’ for his opponents to learn of their errors, and here and Krugman here, saying Williamson had not explained a story of how the effect he describes derives from individual actions.
I don’t much care for Willaimson’s NK modelling style and scary plumbing. DGSE doesn’t properly account for state, accumulation and stocks – only flows, and the Lucas ‘cash balance constraint’ is just a hack to avoid having to account for stocks. None the less Williamson seems to be moving in the right direction. His latest model has a decidedly ‘non-neutral’ role for money. Lets focus on the key mechanism in his model the changes in liquidity from QE leading to a lessening in the ‘liquidity premium’.
I have never much myself like the term and the treatment of liquidity in for example IS/LM models. If liquidity has a premium it has a price and what therefore determines that price? Quantitative stories alone don’t tell you about the demand side or the dynamics of the liquidity premium in disequilibrium. For example the price of an asset at a term will be determined by the next most beneficial asset at that term. The ‘liquidity premium’ is the difference in price with a comparable asset which is fully liquid. But we cant solely determine that price on the basis of a liquidity transformation of that asset only, the price is set by the price of the next most beneficial asset with that liquidity. Why do we want a more liquid asset? Aside from the transactional demand for money it is likely that what looked a good investment is now suboptimal, and in really bad times with negative returns money is the best asset of a bad lot. Seen in these terms we can tease out the differential components of ‘liquidity premiums which are easily confused, rate of returns on assets, the effects of inflation and the effects of risk.
This is how I see the basic macro, in terms from the godfather of QE Richard Werner.
Financial Assets Real Assets(GDP)
Nominal Rate of Return X inflation premium X risk premium <> Nominal Rate of Return (after depreciation) X inflation premium X risk premium
This sets the bounds within agents react and the context for micro, the pages on which the story is written. The inflation premium is the same on both sides of the equation. However the effect is not inflationary neutral, the adjusted rate of return on the RHS is what monetary theorists long ago (after Henry Thornton) used to term the ‘mercantile rate of return’ and on the LHS the ‘monetary rate of profit’. If the rate of change on the RHS is greater than the left hand side then the circulation of goods will parri passu not be backed by the creation of money, i.e. inflation. If the converse happens we have deflation. Where we have a differential rate of return then we see a shift in investment between the two sectors (as in the Petty/Ricardo/Malthus Silver-Corn model I discussed here) leading to an increase/decrease in the real wage which restores equality in the long term providing there is no change in the stock of money in circulation. Changes in the monetary stock, such as paying off debt or increasing the stock of money in circulation, can set off a ‘cumulative process’.
I have previously argues, for example in my paper on a stock-flow consistent alternative to IS/LM that the concept of liquidity could be fully replaced by conceptions of risk and leverage. That is correct and can explain why the ‘demand for loans’ curve (leverage) slopes down with the interest rate. But there is a role still for ‘liquidity premiums’ considered as an addition to the gradient of this curve caused by increasing risk (and not other effects) through time. Consider the following from Anderson’s Economics and the Public Welfare
“Selling on the stock exchanges at the outbreak of the war was an illustration of a fundamental principle in economic life. When there is general confidence in the uninterrupted goings on of economic life, confidence in the legal framework under which economic life operates and in the essential integrity and fairness of governments, men with capital prefer to have their capital employed. They want income from it. They want capital to work with, as giving additional scope to their personal efforts and their personal abilities. They are quite content to have their capital embodied in physical goods destined for future sale, in shares in industrial undertakings, in real estate which brings in rentals, or in loans to active men engaged in industry and commerce. But when grave uncertainties arise, and, above all, when unexpected war comes, men prefer gold to real estate. The man who has his wealth tied up in lands can make no shift. He must sit and take what comes. With the apprehension of war. however, the effort is made to convert illiquid wealth into liquid form as rapidly as possible, even though heavy sacrifices are involved.”
Jerry Bower calls this the Szlazard principle after the physicist who always kept packed suitcases at the door. At times of greatest uncertainty there will be a marked increase in the ‘liquidity’ premium.
Consider again our equation. If the risk premium on financial assets is reduced, through for example a central bank buying bonds on the secondary market, then this will underwrite the risk free return and lower the liquidity premium. This means the LHS value will increase compared to the RHS which is deflationary. We have a simple mechanism, will this satisfy the critics?
A potential objection is this a revival of the ‘natural rate’ argument which Sraffa has supposed to have disproved. He did not only that under barter there is no unique natural rate. See this paper here, Own rates, in any inter-temporal equilibrium, cannot deviate from each other by more than expected price appreciation or depreciation (including the effect of the change in the purchasing power of money) plus the cost of storage and the service flow provided by the commodity, multiplied by the liquidity premium must be equal, so that the net anticipated yield from holding assets are all are equal.
Putting this in broader context of whether the state of ‘mainstream’ economics is in good shape, as Krugman and Williamson have argued. Noah Smith
what Steve and I usually argue about is the general state of macro – he says macro is in fine shape, I say it hasn’t discovered much. I think this reversal supports my thesis. If a top-flight macroeconomist, who knows the whole literature backwards and forwards, can so easily change his workhorse model in one year, and reverse all of his main predictions and policy prescriptions, then good for him, but it means that macroeconomics isn’t producing a lot of reliable results.
But there is nothing wrong with changing your mind, as Keynes did, if a short term crisis leads to dismantling of a misguided ‘mainstream’.