Liquidity trap? Aha!
A quick historical divergence: it's much more useful to study the 19th Century recessions to see this, as at that point M, money, was determined by the supply of gold. And yes, recessions really were accompanied by significant deflation and also falls in output and no, no one really liked that at all. Which is why this idea that Bitcoin has only a single total amount that will ever be issued is not to the taste of central bankers. Because it ties hands when dealing with this specific problem.
And then at this point in comes our Hicksian liquidity trap. So, we've reduced interest rates in order to gee up V from its now lower level. But now interest rates are at zero and we're still seeing a falling total (ie, M3 or M4) money supply and Ooops! Whadda we gonna do? Because we still don't want falling either P or Q. Or PQ come to that.
The obvious answer is let's boost M. And we do that through quantitative easing. The central bank just invents some money, uses it to go out and buy assets (largely, the government's own bonds but it doesn't have to be) thereby boosting the amount of M0 and M1 in the economy. Our transmission mechanism, V, may be broken or partially malfunctioning, but we can overcome that simply by flooding the place with M, so as to avoid that fall in PQ. (Technically, buying government bonds lowers the interest rate on risk free assets, making people go out along the risk curve to buy corporate bonds, stocks, whatever, in order to turn a penny. This has the same effect as lowering the cost of capital to companies and that's what we want, more companies to be willing to borrow to do something.)
At which point we've our first answer to why QE didn't create inflation: it did, but by stopping deflation, not by causing prices to actually rise.
We can also answer one of the other questions floating around: but if the money didn't actually reach the consumer economy then why did we do it? Because we didn't want the money to reach the consumer economy. If we could get money through, if that relationship between M and V had held up we wouldn't be having the problems we were. We're not trying to get M4 to rise enormously (which is the same as saying feed through into the consumer economy) we're trying to get M4 to stop falling, by overcoming the malfunction in V.
And this also, of course, gives us our second answer, that this time around the central bankers actually got it right. And the reason they got it right was in fact Milton Friedman.
His magisterial work, with Anna Schwartz, A Monetary History of the United States, made exactly this point about the actions of the Federal Reserve in the Depression. They allowed the money supply to shrink by not realising that V had collapsed. Ben Bernanke was a student of this period, this subject, and has actually announced to the ghost of St. Milt along the lines of “Milton, you were right, we got it wrong last time and we're not going to this time”.
The BoE agreed, so Britain and the US have come out of this a lot less badly than they could have done. The Bundesbank, and therefore the European Central Bank, did not believe all this and it's one reason why the Eurozone performance has been so much worse.
So, so far, QE hasn't caused inflation precisely because the money hasn't flowed though into the wider economy. It's only stopped, or at least curtailed, a fall in that wider money supply. Which – given that that's what it was designed to do – is great.
All of which leads us along to two further points.
The first is that at some point V is going to return to its former value. But we've now got an extra £375bn of M1 flowing around the system, meaning that when V does return we'll have a vast increase in M4 and that's when the inflation will come. So, we need to be able to reverse our QE: and everyone is planning to do so.
Easy enough to do, just sell the assets, the bonds, back into the market and cancel that base money received for them. But this also has an interesting side effect. There are those who say that the national debt is very much lower than headline numbers suggest. Because the BoE, an arm of government, owns that £375bn of government debt. Heck, why not just cancel it and we're home scot-free?
Why not? Because as and when V returns to normal we need to be able to pull that base money out of the system to stop a jump in P. Both the Fed and the BoE have indicated that they're actually unlikely to sell off their portfolios.
Instead, what they have been doing as parts of them mature (ie, come up for repayment) is to buy more bonds to maintain their stock. At some point, they'll stop doing so: but this means that in order to roll over the bonds (i.e., borrow that amount again) the government will now have to sell more bonds to the markets. This has exactly the same effect as cancelling that extra M1 that was originally issued.
And then there's the entirely fascinating point that many seem to be ignoring from all this. Krugman, for example, keeps doing the victory laps and mentioning Hicksian whatevers. And absolutely correct, conventional monetary policy doesn't solve these sorts of problems, because we hit interest rates at zero, can't lower them any more and that's what the point being made is. And this is where the Keynesian analysis of recessions really starts.
So, conventional monetary policy cannot work, that's when we go off into fiscal policy. Politicians get to spend more than they're taking in in taxes because that liquidity trap means that monetary policy can do no more. Hurrah, eh? And as I've mentioned here before, those who would prefer a larger state quite like this idea, too.
But our recent experience shows that while conventional monetary policy doesn't work, our attempts at unconventional monetary policy seem to have done rather well. Very well in fact. We've just conducted a very large scale experiment into whether, at that zero interest bound, the only possible solution is fiscal policy, deficits and ramping up the spending. To which the answer is no.
It's still entirely possible that borrowing more money for the tax leeches to spend on our behalf is a better way of dealing with this problem (although people like Scott Sumner tell us that it won't be, because the central bank will just change monetary policy to accommodate whatever fiscal policy is) but it is quite obviously no longer the only way. At which point of course the inevitability of Keynesian deficit spending in a recession rather goes away, doesn't it? Still room for it to be better but it's not our one and only solution possible.
And that, I think, in the long term is going to be the biggest intellectual change in macroeconomics. OK, it's going to take a couple of decades for everyone to change favourite hobby horses (perhaps even on the basis that science advances one death at a time) but I do think that in a few decades the conventional wisdom will be that monetary policy, unconventional if need be, can handle recessions just fine. Not perfectly perhaps, maybe there's still a place for fiscal policy, but the insistence that only deficit spending can help will die.
Simply on the basis that this is the first time anyone really tried that unconventional monetary policy, that QE, and it does seem to have worked. This recession would most certainly have been a great deal worse without it, as the contrasting fate of the Eurozone shows.
It stopped the liquidity trap from causing deflation or a worse contraction in GDP, as such we can say it created inflation: and that's what it was meant to do, it worked. Or, as we might ponder on the Keynesian standard solution: what need have we for the politicians to put the next generation into debt when we can sort it out just by inventing a bit more money today? ®