(Originally published in the business pages of the Melbourne Age and Sydney Morning Herald on 21 April 2010)
One of the sillier propositions which has been propagated on the internet and in a range of investment newsletters over the past couple of years is the idea that the ‘quantitative easing’ strategies pursued by central banks such as the US Federal Reserve or the Bank of England in response to the financial crisis amount to ‘printing money’ in order to finance mushrooming budget deficits, and must inevitably lead not merely to higher inflation at some point in the future, but (in the more extreme variants) to the sort of hyper-inflation experienced in Weimar Germany in 1923, or more recently in Robert Mugabe’s Zimbabwe.
It’s certainly true that central banks in most of the major advanced economies have pursued a variety of unorthodox strategies since the onset of the financial crisis in order to provide liquidity to the banking system, to support particular financial institutions or markets, and to get around the problem created by the inability to reduce interest rates below zero. These strategies have generally entailed central banks making loans or purchasing securities that they ordinarily would not, in much larger amounts than they ordinarily would, and for longer periods than they usually do.
Before the onset of the financial crisis, the US Federal Reserve’s balance sheet was of a size in the order of about US$900 billion, with the vast majority of its liabilities comprised of US dollar notes on issue, and most of its assets being holdings of US Treasury securities (that is, debt issued by the US Federal Government). After the collapse of Lehman Brothers in September 2008, however, the Fed’s balance sheet expanded rapidly, to around US$2¼ trillion (US$2,250 billion), by November 2008, and has remained at around that level ever since.
At no time since the collapse of Lehman Brothers has the Fed had more US Treasury securities on its balance sheet than the US$790 billion it had in mid-2007, when the financial crisis initially erupted. As of early this month, its holdings of US Government debt stood at just under US$777 billion. Thus, it is simply not true to say that the Fed has financed, or ‘monetized’, any part of the increase in the US Budget deficit since the onset of the financial crisis.
Rather, the money ‘created’ electronically by the Fed has gone largely to provide liquidity to parts of the US or global financial system that were critically short of it. For example, beginning shortly after the collapse of Lehman Brothers it provided as much as US$500billion worth of US dollars through swap facilities with foreign central banks (including the Reserve Bank of Australia). These facilities have now closed. The Fed provided liquidity to the US banking system and to various markets within the broader US financial system (such as that for commercial paper) in excess of US$1,000 billion on several occasions late in 2008. This support is now down to less than US$60 billion. It also provided over US$100 billion for the bail-outs of Bear Stearns and AIG; most of this is still outstanding.
Beginning early in 2009, and especially since the middle of last year, the money created by the Fed has been directed towards propping up the American mortgage market. As of early this month, the Fed holds over US$ 1 trillion of mortgage-backed securities, as well as almost US$170 billion of debt issued by Fannie Mae and Freddie Mac, the two giant US mortgage insurers now officially in what, in Australian parlance, would be called ‘administration’. Prior to the collapse of Lehman Brothers, it held not a dollar of either; they now account for almost half the assets on the Fed’s balance sheet.
Without this support, the mortgage market would have been in even more difficulty than it was, and American house prices would presumably have fallen by even more than the 30% which they did from their peak in mid-2006 to their trough in June last year.
The increase in the Fed’s holdings of mortgage-backed securities and debt issued by Fannie Mae and Freddie Mac is matched, almost exactly, by an increase in the cash held by the commercial banks in their accounts at the Fed, from typically less than US$10 billion prior to the collapse of Lehmans to over US$1 trillion since last October.
What happened, in other words, is that the Fed has injected over US$1 trillion into the market for mortgage-backed securities; the sellers of those securities deposited the proceeds, directly or indirectly, with their banks; and those banks have held the cash on deposit with the Federal Reserve (as opposed to lending it out again).
There’s simply no way that this can be inflationary. Inflation would only become a risk if the Fed failed to unwind the expansion in its balance sheet once the banks start to lend out the funds which they are currently holding as reserve balances with the Fed. But the Fed has made it very clear that they are aware of this risk, and have both the inclination and the means to deal with it when it arises.
Meanwhile, the burgeoning US budget deficit has been financed – rather smoothly, judging by the absence of sustained upward pressure on US government bond yields or downward pressure on the US dollar – by sales of Treasury notes and bonds to households, who have been saving more and borrowing less since the onset of the financial crisis (and whose direct holdings of US Treasury securities have risen by almost US$400 billion since the collapse of Lehmans); to American banks (whose holdings of Treasuries have risen by some US$300 billion since the collapse of Lehmans); and especially foreigners (whose holdings of US government securities have risen by more than $1,000 billions since the collapse of Lehmans).
The US Federal Reserve hasn’t been the only central bank pursuing these ‘unorthodox’ strategies. The Bank of England, in particular, has expanded its balance sheet by slightly more (relative to the size of the British economy), than the Federal Reserve. And the balance sheets of both the European Central Bank and the Bank of Japan are larger, relative to their respective economies, than those of either the Fed or the Bank of England.
Japan’s experience is particularly instructive in this regard. The Bank of Japan has been consciously trying to engineer a positive inflation rate for the best part of a decade: and yet in only one year of the past ten has Japan’s ‘core’ inflation rate not been negative.
The example of Japan shows that, in circumstances where supply exceeds demand by a wide margin – as is the case in most of the major advanced economies at this time – creating inflation is actually quite difficult.
This is the exact opposite of the examples commonly cited by those drawing analogies with Weimar Germany or Mugabe’s Zimbabwe. In those and other instances (including Nationalist China under Chiang Kai-Shek, Germany and Hungary immediately after World War II, many Latin American economies in the past five decades, and much of the former Soviet empire after 1989), the ‘supply side’ of the economy had collapsed as a result of enemy occupation, wartime destruction, institutional collapse or years of egregious economic mismanagement. And in those circumstances, copious money-printing by the central bank in an attempt to sustain demand inevitably led to massive inflation. But those circumstances are a world away from those confronting the world’s major advanced economies today. And the parallels which are sometimes drawn with them are utterly spurious.
(Saul Eslake is a Program Director at the Grattan Institute. The views expressed here are however his own).
Hi Saul,
Interesting piece. Acouple of small points. You make a sharp distinction between the actual current decisions of the Fed Reserve and the actions of the Federal government, making it possible to say that the increase in debt of the Federal Government via foreign holders is not an inflationary pressure that can be blamed on the Federal Reserve. It does in principle add inflationary pressure, but simply not due to the Reserve. Many commentators wouldn’t make such sharp distinctions partially because they would, either implicitly or explicitly, think that when push comes to shove the Federal Reserve would roll over Government bonds if no-one else did, and hence partially blame the Reserve for allowing the money supply to be increased via borrowing of the Government.
Second, if you want to argue about the effect of the loans to banks on inflation, its no good just talking about the amount of money banks hold at the Reserve without looking at whether this has crowded out other securities banks could have held and hence whether it hasn’t indirectly increased their lending activities (these loans were certainly meant to get banks lending again!). I am not going to put in the effort to see if this has happened, but will note the possibility.
Thirdly, since the Fed is ultimately in the position to get the money supply to whatever level it wants, you can in principle blame it for the level of money supply out there and hence look at the levels to see whether it has ‘allowed’ monetary expansion. If you look at M2 for the US (http://www.federalreserve.gov/releases/h6/hist/h6hist1.txt), you see it has increased by about 700 billion since October 2008, which is close to 10%. That’s quite an increase for a country in recession, but since the velocity of M2 also seems to have gone down, this monetary expansion hasn’t yet lead to the inflationary pressures envisaged.
Of course you are right that the analogy with classic money printing is imperfect, but on the other hand there are now ways of increasing the money supply without direct printing that were not available to the Weimar republic. I guess it would be handy to know which statements by which commentators you directly want to take a shot at.
So in summary: the Federal Reserve bought mortgage backed securities and paid in cash. They have not actually paid (yet) in real cash cash because the sellers of the securities don’t have anything useful to do with the money so they are content with an electronic promise that the Federal Reserve will stump up said cash at some unknown future time.
Unwinding would require selling off the same mortgage backed securities for something close to what was paid. Given the 30% fall in value of the houses sitting underneath these securities (and the continuation of mortgage defaults to the point where the banks don’t even talk about it anymore) you must be expecting a buyer to fly in from the North Pole with a big white beard and a red hat.
Thank you for being so reassuring. I notice that you hesitate to elaborate on any details nor do you provide references, but on a hand waving guess I presume we are talking about those submarines full of gold bars on their way into China. :-) Sure, this deals with the inflation problem… temporarily… until neither Fort Knox nor the Comex have any metal left in them.
Dunno if this is in any way relevant…
http://www.businessinsider.com/hidden-in-the-ppi-data-was-the-largest-food-price-spike-in-26-years-2010-4
In response to the first of Paul Fritjers’ “small points”, I wasn’t really seeking to draw a distinction between the Federal Reserve and the US Federal Government as alternative sources of potential inflationary pressure, although I can see how that interpretation might be drawn from what I wrote. In that context, I do not think the budget deficits currently being incurred by the US government are themselves inflationary in circumstances where the ‘output gap’ between potential supply and effective demand is so large. Rather, had the government been unwilling or unable to go into the deficit on the scale on which it has, the downturn in overall economic activity would have been even greater. As with the expansion in the Fed’s balance sheet which was the subject of my article, the risk of inflation would become more tangible if the budget deficit were to remain large even as the economy recovered sufficiently to close the output gap. I have rather more confidence in the Fed’s willingness and ability to unwind its expansionary monetary policy in a timely and orderly way than I do in the Administration’s and Congress’ willingness and ability to do the same with fiscal policy.
To Paul’s second point, I think the increase in banks’ holdings of cash in their accounts at the Fed is a reflection of a combination of their reluctance to lend (either directly or via the purchase of debt securities) and weak demand for credit from the private sector. As yet I don’t think there is any evidence of ‘crowding out’ by the Fed’s purchases of mortgage-backed securities. Indeed, if and when private investor demand for those securities begins rising, that will itself serve as a market signal for the Fed to start offloading its holdings of them.
By my calculation, M2 has risen by 7.9% between September 2008 and February 2010 which as Paul says is considerably faster than nominal GDP (less than 1%) so velocity of money has indeed declined – as one would expect in a recession, particularly one in which financial dislocation has played such an important role. Again, the Fed needs to be alert to the timing and magnitude of the eventual recovery in velocity when calibrating its ‘exit strategy’.
Among those who have compared the Fed’s ‘quantitative easing’ strategies with those of Weimar Germany or Mugabe’s Zimbabwe are Marc Faber (‘US Inflation to Approach Zimbabwe Level, Faber Says’, Forbes, 2 June 2009) and Bill Gross (‘Gross Says Diversify from Dollar as Deficits Surge’, Bloomberg, 3 June 2009), as well as any number of gold-bugs, self-proclaimed ‘Austrians’ and others on the internet and among investment newsletter writers.
These references may also be of interest to ‘Tel’ who berates me for not providing any; I can only say that this was originally an article for a newspaper, and it’s not usual to provide footnotes in newspaper op-ed pages (especially those subject to word limits). Whatever gold is in Fort Knox will stay there; the US dollar hasn’t been convertible into gold since 1971.
I think it’s a long bow to draw between the Fed’s ‘quantitative easing’ and the large rise in the PPI in March, reported on Friday night our time. As ‘Tel’ notes, this was largely driven by a sharp rise in the (highly volatile) food component. The overall PPI rose 6.1% over the year to March, but its absolute level is still below where it was in September 2008. Meanwhile the ‘core’ PPI (excluding food and energy) rose by just 0.1% in March to be only 0.8% higher than a year earlier. More generally, much of the movement in producer prices reflective changes in relative prices – in particular, increases in commodity prices driven in large part by demand from emerging markets – rather than in the overall price level (including services). Consumer prices excluding food and energy rose by only 1.2% over the year to March. Not much sign of inflation there.