“we are able to catch a first glimpse of the way in which changes in the quantity of money work their way into the economic system. If, however, we are tempted to assert that money is the drink which stimulates the system to activity, we must remind ourselves that there may be several slips between the cup and the lip. For whilst an increase in the quantity of money may be expected, cet. par., to reduce the rate of interest, this will not happen if the liquidity-preferences of the public are increasing more than the quantity of money; and whilst a decline in the rate of interest may be expected, cet. par., to increase the volume of investment, this will not happen if the schedule of the marginal efficiency of capital is falling more rapidly than the rate of interest; and whilst an increase in the volume of investment may be expected, cet. par., to increase employment, this may not happen if the propensity to consume is falling off. ” J.M.Keynes 1936
This is my second post about quantitative easing (QE). This subject may seem arcane but I think it is crucial to the general theme of this blog. As I discussed in that previous post, QE is being held out as a painless solution for a stagnant economy. As such, any other policies to sort out our economy get put to one side in favour of giving QE more of a chance. Meanwhile QE seems to have somewhat mysterious effects on asset prices and those impact on inequality and capital flows between countries and so effect us all.
I think it is important to make a clear distinction between simply using QE to bring risk free interest rates down and using excess QE, above and beyond that, in an attempt to satiate any preference the public may have to hold wealth in the form of zero yielding cash rather than as other asset classes such as stocks or bonds. In WWII and its immediate aftermath, the US and UK governments both used QE to peg low interest rates for both short term and longer term government bonds so as to facilitate massive deficit spending. That is a straightforward policy and keeping such low interest rates could be a permanent arrangement much as it has been in Japan for the past couple of decades. However we now have QE being used not so much to facilitate government spending but as an alternative to that as a way to try and stimulate the economy.
In my previous post I tried to rationalize the effects of further QE (once risk free interest rates were at the lower bound) in terms of how it could place a ceiling, for a period, on the extent to which the central bank could subsequently raise interest rates. Nick Edmonds has expressed skepticism as to whether such a constraint would actually come to bear on a central bank that wished to hike interest rates. His point is that, one way or another, a fix would be found.
Nick Edmonds models QE in terms of how purchases of assets such as government bonds (by the central bank, from the public) induce price changes in all other asset classes, domestic and foreign, as the public rebalances holdings of those various asset classes. This also tallies with how the Bank of England explains QE effects. To me this raises the crucial issue of why “the public” chooses to exchange other liquid assets for cash* even when cash is a zero yielding asset and the assets they sell do have an appreciable yield. In principle in the simplest model, a liquid asset with ANY yield would be preferable to zero yielding cash that was in excess of what was needed as a medium of exchange. Such a view however overlooks the crucial attribute of cash as the ideal stable store of value over the short and medium term (1). Perhaps the clearest description I have seen for this is from an interview with Warren Buffett’s biographer Alice Schroeder:-
“”He thinks of cash as a call option with no expiration date, an option on every asset class, with no strike price.” It is a pretty fundamental insight. Because once an investor looks at cash as an option – in essence, the price of being able to scoop up a bargain when it becomes available – it is less tempting to be bothered by the fact that in the short term, it earns almost nothing. Suddenly, an investor’s asset allocation decisions are not simply between earning nothing in cash and earning something in bonds or stocks. The key question becomes: How much can the cash earn if I have it when I need it to buy other assets that are cheap, versus the upfront cost of holding it? “There’s a perception that Buffett just likes cash and lets cash build up, but that optionality is actually pretty mathematically based, even if he does the math in his head, which he almost always does,””
Asset managers selling assets to the central bank obviously must value the cash they receive just as much as the assets they sell. This undermines the popular narrative about QE causing a “hot potato effect” as asset managers try to offload zero yielding cash. More generally whenever anyone sells an asset to anyone, it is because they value that amount of cash as much as that asset. That, after all, is what sets the market price for assets. So every asset is valued as that amount of “call option …with no expiration date… on every asset class, with no strike price” to use Alice Schroeder’s description of cash in its asset class guise.
Every asset also has its price set by what it itself is likely to earn. Peters and Adamou make the case that asset prices and price volatility find levels at which borrowing money to purchase more of an asset won’t increase gains. Buying more of an asset using borrowed money will amplify any short term gains but also any losses whenever the asset price falls and those losses will outweigh the increased gains overall. This stems from the phenomenon of “volatility decay”. A 25% fall in price requires a 33% rise to get back even,whilst a 50% fall requires a 100% rise. This effect would cap any durable upward repricing of assets unless price volatility also reduced and/or earnings increased.
Obviously, just as cash can be used to “scoop up a bargain when it becomes available“, so any other liquid asset can potentially be sold to fund such a purchase. The asset manager’s ideal would be to have holdings of diverse asset classes that had prices that could be relied upon to move in opposite directions such that bargains in one asset class could be bought when the other asset holdings had boosted available funds. The asset value of cash relies on the unreliability of such negative correlations between the price movements of various assets. I suppose this is making the same general point as J.M.Keynes’s statement :-
“It might be thought that, in the same way, an individual, who believed that the prospective yield of investments will be below what the market is expecting, will have a sufficient reason for holding liquid cash. But this is not the case. He has a sufficient reason for holding cash or debts in preference to equities; but the purchase of debts will be a preferable alternative to holding cash, unless he also believes that the future rate of interest will prove to be higher than the market is supposing.” In this context J.M.Keynes means future interest rates on long term debt or in other words, bond prices.
Of course cash doesn’t simply serve as an asset class; its value is also anchored in the real economy in terms of how much goods and services it can buy. Cash in the hands of asset managers has little influence on the prices of real economy goods and services. Asset managers simply work to transport value through time and use wealth to gather more wealth by buying and selling securities. The cash they control is restricted to that use and so does not generally take part in bidding for goods and services.
So basically the financial system is dealt exogenous constraints from the real economy. Cash has its real economy value and the firms and property underlying assets can only supply asset holders with gains limited by real economic activity. The financial system then juggles that around amongst itself. For everything to still tally up, the extra cash held by the public due to QE must shift the return characteristics of the market to justify the continued exchange of yielding assets for yet more cash. The Peters and Adamou study indicates that increasing levels of cash are unlikely to durably push up asset values. Those selling assets to the central bank also clearly have that perception or presumably they wouldn’t sell. It seems to me that something has to give and perhaps the most likely effect is to synchronize the price movements of the various asset classes. Perhaps that unfolding scenario is the only reason why the public continues to choose to part with assets in exchange for zero yielding cash.
All the cash provided by QE has to be held by someone at all times as with all other issued securities until they are withdrawn. Prior to all of the QE, an upward revaluation of say mining stocks might have entailed a sell off of say long term treasury bonds so as to provide the funds. The last participant to sell his treasury bonds would not have got much money for them and would not have been able to buy many shares of the mining stocks. By contrast, in a post-QE world, the mining stocks could be bid up by buyers who did not need to sell off anything else because they had ample cash reserves to draw upon. Consequently all assets now can be bid up in unison and likewise they can all fall in unison. The more that effect takes hold, the more valuable cash becomes as an asset class. In June 2013 there was a modest drop in asset prices that nevertheless was widely remarked upon because it seemed perfectly synchronized across all major asset classes; all classes of stocks, bonds and commodities, across the globe, seemed to suddenly revalue downwards in terms of the US dollar or Sterling.
Its worth asking whether this is a problem or not. If asset managers now hold more cash and need to hold more cash, then perhaps that is a fairly neutral non-effect. However, at the very least, it makes me very much doubt that we can sort out our economic problems simply by doing more QE. To me the main downside to QE looks to be a further dislocation between the financial system and the real economy. In an ideal world, asset markets would serve the real economy by coordinating allocation of real resources towards where they could best serve the real economy. I worry that QE might only serve to make asset markets yet more stochastic, disruptive and capricious in their effects on the real economy. Perhaps the starvation that came for many people as a result of the seemingly stochastic spikes in grain prices in 2008 should serve as a warning of the potential real damage irrational financial markets can cause.
notes:
*I’m using “cash” here in the sense of any “cash equivalent” such as risk free bank deposits, very short term treasury bills, and, for banks, bank reserves.
(1) I think Keynes was alluding this when he described the “liquidity-preference due to the speculative motive”. For me however, Keynes’s terminology is unfortunate because this is all about the price stability of cash, not the fact that cash can readily be exchanged for other assets as the adjective “liquidity” is often used to mean.
Related Previous Posts:
Monetary policy, the 1930s and now
Larry Summers at least sees the problem
Related stuff on the web:
Money creation in the modern economy -Bank of England (link added 19march2014 ht JKH)
For Warren Buffett, the cash option is priceless -The Globe and Mail
Chapter 13. The General Theory of the Rate of Interest -J.M.Keynes 1936
Stochastic Market Efficiency -Ole Peters and Alexander Adamou
The Case Against Monetary Stimulus via Asset Purchases -Ashwin Parameswaran
What the *&%! Just Happened -Ben Inker -GMO
Leash the Dogma -John P. Hussman
Investing Like the Harvard and Yale Endowment Funds -Frontier
direct economic democracy 
On the question of comparing yields on cash with those on bonds and equity, you have to compare expected rates of return over the same time period. So for the short term, for cash (or bank money), that’s just the rate paid. For a bond, it’s the current yield, plus or minus the expected change in price over the period. The theory for why people are prepared to hold cash at such low rates is that they expect similarly low returns on bonds and equities, because they expect the price to fall. Which is consistent with the price of those assets being high.
Nick, isn’t it sensible though for asset managers to always have a strategic asset allocation to cash based not so much on their presumption that bonds and equities are currently over-priced but rather on the presumption that price movements will bob around in a way that they wont be able to predict? So an asset manager might choose to hold say 40% stocks, 25% long term treasuries, 10% cash, 5% commodities and 20% land and keep those allocations for years through several market cycles. Then QE happens and all the volatile asset prices start moving in the same direction rather than in different directions. So the original asset allocation strategy basically stops working, so the asset allocations are revised to allow much more cash. Now the ideal portfolio might be say 30% stocks, 20% long term treasuries, 30% cash, 5% commodities and 15% land or whatever. The asset allocator is basing this entirely on the risk of all of the asset classes falling in price together. The asset allocator could be entirely agnostic as to whether prices are more likely to rise in the coming period or fall in the coming period and considers such market timing to be a hopeless fools’ errand. I put the link up about the Harvard and Yale endowment funds having very stable and diverse asset allocations right through several market cycles. Same thing with that Ray Dalio “all weather portfolio”. They are making no attempt to market time and are huge funds and are widely copied aren’t they?
Even aspirant market timers such as GMO don’t go all in on the basis of their predictions. They may consider that say US stocks are currently overvalued and so hold more Indian stocks or whatever but they are only tweeking the percentages. They don’t sell up everything and go 100% to the other asset class based on their prognostications. If anyone could make reliable prognostications, then it wouldn’t take long before they owned all global assets. Bold prognosticators come to a sticky end don’t they and even timid prognosticators such as GMO don’t outcompete the more agnostic asset allocators.
I think it is perfectly possible that QE could have no influence at all on the long term total returns from assets and yet cause rational asset allocators to choose to hold more cash purely on the basis of QE causing prices of different asset classes to tend to fall in unison whenever they fall.
Let’s imagine the aim was to have a perpetual endowment fund that paid out as much as possible per year without being able to pay out more in good years and less when the portfolio returns had a bad year. The hypothetical asset manager can only choose between stocks that over the very long term (say 50 years) give total returns of say 8% per year, long term gilts that give 4% per year and cash that gives nothing and only those assets are available. The choice as to how much of each of those assets to hold is governed by the volatilities of the asset prices and of the income streams from them and the correlations between those. Stocks might have a Japanese style multi-decade plummet in both prices and dividend payments. If as in say Japan in 1990-2002 that coincided with a huge surge in long term treasury (JGB) prices, then having those would save the day. BUT if both assets plummeted in price at the same time, then cash would be vital. If an asset manager just held stocks then they would be risking calamity because sooner or later the dividend stream could (temporarily) dry up and the stock prices fall such that they had to sell at just the wrong time inorder to meet pay out requirements to whoever was the beneficiary. That would be a rapid death spiral for the fund.
Yes, most managers running balanced portfolios are not going to switch to 100% cash, they are just going to tweak the percentages a bit. In my own experience though this is much more likely to be based on a view that particular markets are over or under-priced, rather than on a perception that the correlation between asset returns has fundamentally changed. Correlation co-efficients are certainly taken into account, but they tend to be based on analysis over fairly long periods.
There will also be more active funds taking positions based on market views and these are likely to make bigger portfolio adjustments. Remember as well that, whilst QE has a significant effect, it has to seen in the context of other things going on. Over the same period, the supply of government bonds has also risen significantly due to large deficits. All of these things have to be absorbed.
Nick, isn’t it true that over the last few years, QE pretty much matched deficit spending though so the public has been absorbing more cash rather than more treasury bonds?
Possibly, yes. I’m just pointing out that there are big movements in both directions.
Nick when you say, “In my own experience though this is much more likely to be based on a view that particular markets are over or under-priced,”
-what puzzles me is why they then choose cash to hold rather than some other asset. Are you saying that they predict that ALL non-cash assets will have negative total returns and that that isn’t that the same thing as saying that they will fall in price in unison? So this year say bonds will fall, then next year stocks will fall and in the interim those prices won’t recover so even a combined and rebalanced stock and bond and cash portfolio will underperform cash?
If the people holding cash are pessimistic about all assets and yet the asset prices are still high, then doesn’t this amount to QE causing some people to be optimistic about ALL asset classes and other people to be pessimistic about ALL asset classes? To me that seems to be getting back to QE causing positive correlations between asset price movements?
If we ignore risk premiums and things like liquidity returns, then financial theory tells us that the overnight return on all assets must be equal. So if cash is yielding 1% and the coupon on bonds is 4% and the dividend yield on equities is 3%, then bond prices will be expected to fall 3% and equity prices 2% (all as per annum figures, so say 1/365ths of that per night). But then we have to factor in risk premiums and liquidity premiums and so. These depend on the quantity held. The more cash you hold, the lower the marginal liquidity premium; the less balanced your portfolio the more risk.
If people think the price will rise they will buy more of the asset and the price will rise. If the price of an asset is high, people may expect it to fall.
Nick, that Ole Peters stochastic market efficiency paper though says that it is a big improvement over the traditional “risk premium” way of viewing asset pricing. Am I misinterpreting it? I don’t know what your (or general) opinion of it is. To me it was an asset pricing theory that seemed to make some kind of sense unlike the normal theory.
There is no mysterious risk premium. All securities are just whatever price causes long term returns to not be improvable by leveraging up.
I’m not familiar with that paper and it would take me a little time to work through. But, I’m being very general about risk premiums, liquidity premiums, etc. I’m just saying that I don’t think the expected return on all assets should be strictly equal, because you have to take other things into account. I think most people would agree with that though they might disagree on what to take into account and how important it is.
I totally agree that the expected return is not expected to be equal and that Ole Peters paper isn’t claiming that. If an asset is very volatile, then it could have great long term returns and yet leveraging up wouldn’t provide any benefit because the increase in each dip in price due to the leveraging would outweigh the gains. That is pretty much in agreement with this too:
http://www.scribd.com/doc/73349539/GMO-Rethinking-Risk-Beta-Puzzle-1112
I have to stress that my maths is far from being up to appraising the Ole Peters stuff in anything more than the most rudimentary intuative makes sense kind of way.