The 'Wall Of Money': A guide to QE2
The US Federal Reserve is set to launch a second round of quantitative easing, known as QE2, on 3 November 2010. It is probably the US policy community's last shot at averting a double-dip recession and it may work. But there is an argument raging among economists over the dangers. Here is a brief outline of what they are doing and why, and the arguments for and against. If you can improve on it, fire away and suggest changes; ditto if you disagree. I've talked to a number of financial sector economists to try to get this right, but it's still a think-piece rather than definitive.
Oh, and the whole future of the world economy depends on who's right.
1. What is QE?
QE involves the central bank creating money and using it to buy up pieces of paper known as bonds. These are IOUs issued by government, companies or banks, so it is effectively the central bank printing money and lending it to the government or others, with the aim of boosting spending power in the economy. First tried by the Fed in April 1932, it was also deployed by Japan in 2001. Both Britain and the USA adopted the policy in March 2009.
2. How much have they done already?
In the USA, $1.75 trillion dollars; in the UK, £200bn (this latter representing an input equivalent of 12% of GDP).
3. Why are they doing it?
To stimulate growth in a severe recession, you have two macro tools: monetary and fiscal. Fiscal is where you either cut taxes or raise public spending (or both). Monetary is where you cut interest rates in the hope of boosting demand, and in the process, the money supply. Early in 2009 both central banks reached close to zero interest rates, and were forced to consider unconventional measures to stimulate the money supply. QE is the key unconventional measure available.
4. How does it work?
We don't know.
4a. Come again?
The advocates of QE - the so-called monetary Keynesians who see April 1932 as the key turnaround moment in the 1930s -- thought it should work like this: you pump money into the banking system until the risk of the banks lending falls. Your first effect is to reduce the cost of borrowing for the government - the so-called risk-free rate; the second effect is to reduce the difference between that and the real cost of borrowing.
However this did not happen. Or rather it only happened for big business - not for small businesses and consumers.
The money got locked inside the banking system. It raised prices on the stock market, and for banks. And it also probably kept house prices from falling further. This is the so called "wealth effect". Also it allowed big companies to "become their own banks" - issuing their own bonds as a way of borrowing, rather than using the commercial banking sector. However the money never flowed through into final demand - small business loans, higher credit card limits, mortgages, cash in the high street. Or not in large quantities anyway.
There is even a flow-diagram produced by the Bank of England to show how it thought it might work: but the flow is stuck somewhere between the wealth effect and the boost to final demand. The money supply is only growing at 0.9% Q-on-Q - compared to 9% average for the 10 years to 2008. The Bank was insistent that QE would boost the money supply - it currently has no public explanation as to when the 12% of GDP pumped in flows through into 12% extra demand. Or as the Bank puts it: "it remains difficult to judge with any precision the ultimate impact of asset purchases on nominal demand." (May 2010 Inflation Report)
5. What did the Americans do differently?
Bernanke had already intervened to start buying up $600bn of distressed debts of the state-owned mortgage companies with central bank money, and he always argued QE should be done in this targeted way. He even travelled to the London School of Economics to warn against doing classic QE in January 2009. But the recession turned out worse than expected and he had to do full-blown QE anyway. However the US version of QE always had a much bigger targeted aspect - ie tactical buying of bad debt as well as strategic buying of safe debt, and therefore probably had a more benign impact on the mortgage market.
6. So what's gone wrong?
In neither the UK nor the US has QE led to a concomitant boost in demand, or a rapid growth in the money supply. It did lead to cuts in the real interest rate (between 30 and 100 basis points in the USA, around 100 basis points in the UK according to the resepective central banks). In fact the short-term inflation protected interest rate is now negative in both countries.
But credit conditions did not improve for consumers and small businesses.
In both countries there is a key symptom: the banking system is hampered by bad debt, while the private sector - consumers and companies - are trying to pay down debt, using any boost to credit conditions to effectively shrink the money supply "from below" just as the governent tries to boost it from above.
The USA has an additional problem: a much bigger overhang of mortgage debt: 1.5m repossessed properties, a rising foreclosure rate, etc. The UK took micro measures to guard against a wave of repossessions; the US took one big micro-measure - a fund to compensate mortgage lenders against foreclosure - that has now run out.
In short, the pro-QE critics say its failure is due to the unwillingness to nationalise the banks, inflict one-off losses on savers and investors, write off housing and commercial property debts etc. Such critics thing the Fed and BoE should have set a target interest rate - rather than a target volume of QE - ie specified the output not the input. And then printed money and manipulated the banking system until that target rate was reached. For the pro-QE crititics, the Fed needs to do a lot more QE fast. They point out that in the 1930s it was only when the Fed doubled or trebled its initial QE purchases that it had a dramatic impact on real interest rates.
The anti-QE critics say it did not work because it could not work.
6a. Do they get it?
Economists at both central banks have produced learned papers on the subject. The Bank of England (Joyce et al) says:
"Our analysis suggests that the purchases have had a significant impact on
financial markets and particularly gilt yields, but there is clearly more to learn about the transmission of those effects to the wider economy."
Which kind of confirms what I've been saying.
The Fed economists (Gagnon et al) say:
"Based on this evidence, we conclude that the Federal Reserve's LSAP programs were successful at lowering longer-term private borrowing rates and stimulating economic activity. While the effects are especially noticeable in the mortgage market, they appear to be widespread, including in the markets for Treasury securities, corporate bonds, and interest-rate swaps."
So the Fed can claim to a greater extent than the BoE that their particular type of QE worked better in the mortgage market. But they do not demonstrate any lasting positive impact on money supply and final demand.
7. Has it caused inflation?
The UK is certainly seeing persistent above-target inflation, and the anti-QE critics predicted this. However the pro-QE people see this as entirely due to the VAT increases. The Bank of England is in a bit of a bind because of the conceit that it has to construct all its policies around is the inflation target. It has to say QE is just an extension of interest rate policy, with the aim of getting inflation within 1% either side of 2%. It has no target for growth, demand, money supply - only inflation. So this justification is wearing a bit thin - as inflation is on the high side of target. Much was made of the undershoot to inflation when QE was launched, in this explanatory document.
7a. However, in the USA there are warning signs of deflation. That's the qualitative difference between here and there. A certain double dip in the housing market, falling real incomes and inflation at around 1% - which is too low to sustain demand.
8. What's the objection to doing more?
This is where we hit the global macro-economics of QE - obvious to people in the market, completely opaque to the ordinary punter. I will take you through it step by step:
a) QE depresses the value of the currency of those printing the money, and forces the value of other currencies upwards.
b) It also depresses the returns savers get on long-term investments.
c) Therefore, in anticipation of more QE, massive amounts of capital have begun to flow out of the USA to the emerging markets. This is known as the Wall of Money or the "global search for yield"
d) Bad outcome #1: emerging markets accept their own currencies have to rise versus the dollar/sterling: but prices of goods coming from the emerging markets now rise. Companies in the west have two choices: pass this on in the form of cost-inflation or take a hit to profits. We get stagflation. This is the critique favoured by a some people in the UK investment community, and sees the West finally and painfully pay the price for allowing so much of its manufacturing industry to move south and east.
e) Bad outcome #2: currency war, already under way, intensifies as emerging markets resist appreciation of their currencies and this this spills over into trade war. The rebalancing of the globe takes place not within a single global market but a fragmented market of regional trade and currency blocs.
9. Is there a danger of monetisation?
When QE was first mooted, some tabloid newspapers predicted it would bring Zimbabwe style inflation because the central bank would use the money printed not just to buy and hold the national debt but to pay it off. That is known as monetisation.
Generally it leads to high inflation because in monetary theory, if a larger amount of money suddenly starts chasing the same amount of goods, the price of the goods will rise. Some commentators think that the Fed has already begun tactically monetising the debt; others believe it's a matter of quantity - that as long as the amount of Treasury debt is smaller than the amount of money in circulation it does not lead to monetisation. So anything between an extra $500bn to a trillion - even if signalled rather than purchased - could be read as an a signal that the US is about to monetise its debt.
FT commentator Martin Wolf has even begun to advocate a form of monetisation. What is clear is that, if you believe you face deflation, then monetisation makes sense: but it is the ultimate form of beggar thy neighbour - because it signals that holding dollars is a bad thing it is an overt signal to devalue the dollar and an overt act of currency competition.
10. What needs to happen to unblock the logjam?
QE1 stabilised the banking system and global trade; it has also filled big companies' coffers full of cash. But they are not yet prepared to invest it; meanwhile small businesses and consumers still face a credit crunch.
There needs to be a defibrillating moment where - in addition to QE2 - large amounts of bad debt are written off in the private sector - above all in the housing market. Then consumers suddenly get access to credit again and then the big cash mountains of the private sector get thrown into the economy in the form of investment.
But for that to happen somebody has to take losses who has not already taken them. That means the banks: they have to take the big hit on mortgages and commercial property they have refused to take; that in turn hits the government, which in the US and UK has "guaranteed the losses" on hundreds of billions of bad debt for the cost of a few tens of billions.
In other words, the final destructive power of the bad debt in the system has to be allowed to work itself out.
On top of that, once the final cathartic moment is over, the central banks then have to get out of money printing in an orderly way, allowing quite a bit of inflation to avoid choking off the recovery. One way to do this would be to temporarily abandon inflation targeting - to say: we will keep printing money whatever happens to inflation, until growth reaches a set target and stays there. Bernanke has toyed with this, and it will be interesting to see if he keeps the idea alive.
It's quite tricky, basically - and it involves dragging the politicians back into monetary policy making, ending the fiction that it is somehow "apolitical".
QE2 will buy time. But in that time the governments have to act at micro-level to restructure the finance system so that it starts working again.
11. Could we see another "cathartic moment"?
Some bank analysts believe the second downturn in the US housing market, with the threat of deflation, poor numbers in general etc has created a new pool of undeclared bad debts in the US banking system, totalling around $500bn. In addition, the long-postponed accounting for bad debt in the commercial property market has to happen at some point.
Both politically and economically there could never be a "second bail out" in the USA - the restructuring would have to look more like what happened when Yamaichi Securities went bust in 1997 - a state-controlled insolvency process for one or more banks. Or, of course, the Mother Of All Cathartic Moments, the enforced bank holiday of 6-9 March 1933 when Franklin Delano Roosevelt - one year into the Fed's quantitative easing spree - forced 2,000 US banks to close for good, followed by a massive injection of taxpayers' money into the surviving institutions (and of course their breakup using Glass-Steagall).
What we know from the 1930s is that QE was not enough: it had to be accompanied by massive restructuring of the finance system and in any case led to a beggar-thy-neighbour exit strategy, rival currency and trade blocs and an extended Depression for the countries that lost the exit battle because they clung to their old orthodoxies.
Addendum: Another theme you should be aware of is the danger that QE causes a "bond bubble" that then collapses, above all in corporate debt. An amusing, if slightly Newsnight circa 1982-style, video from the FT explaining this click here.