Working in the hedge fund business and investing in the financial services sector, I spend a lot of my time trying to understand the interactions of monetary and fiscal policy with financial markets. In the wake of the financial crisis, governments around the world have undertaken plenty of unorthodox policy measures, with varying degrees of success or perceived success. Recently I've been trying to better understand QE, the nickname commonly applied to quantitative easing by the US Federal Reserve or other central banks.
Lots of ink has been spilled on the subject, though it seems that the ultimate impact of the policy is still not well understood by pundits or market participants. In general there seems to be a consensus that the current round of QE2, started last year, is evidence of ultra-loose monetary policy and that the inflationary pressures we're seeing around the world are the inevitable result. Although I'm not denying that various measures of money have increased sharply since the crisis, blaming this on QE or even on the Fed is an oversimplification or even a misrepresentation. Since QE2 is slated to end in another two months and the implications for markets are considered to be significant, I thought it might be useful to go step by step through the mechanics of QE. The conclusions may surprise you.
One unassailable fact is that the Federal Reserve has dramatically expanded the size of its balance sheet over the past several years. Note that since before the crisis began in 2008 the Fed's assets have just about tripled. As its assets have expanded so, mechanically, has the monetary base. On the surface this sounds like an open and shut case: The Fed has consciously expanded the money supply, and the price of oil and gold, etc. is soaring as a result.
But what exactly is the monetary base, how is it composed and how does its growth impact financial markets? Well, in fact the monetary base is simply the liabilities of the Fed. Historically these liabilities have overwhelmingly consisted of currency in circulation. That's why they're called Federal Reserve Notes: they're the debts owed to their holders by the Fed. In the days of the gold standard you could take your dollar to the Federal Reserve and ask for the equivalent value of gold that backed it. Now try that and you'll get...a dollar. Anyway, the point is that until the crisis happened the monetary base, i.e., the liabilities of the Fed, consisted mostly of cash.
In October 2008, at the height of the financial crisis, the Federal Reserve announced that it would begin paying interest on reserve deposits by financial institutions. These reserves are the second main component of the monetary base, which makes intuitive sense since deposits are an obligation to the depositor by the institution that holds them. Notice what's happend to the level of bank reserves held at the Fed over the past several years: They've gone from almost nothing to about $1.5tn. Meanwhile, currency in circulation has grown by only about $200bn to around $1tn.
Let's step back now and think about what QE2 has meant in practical terms for the various participants in the market for government debt. In the first place, the Treasury issues bonds to the Fed, which obligingly presses a button and creates a credit on the Treasury's account. Now the Treasury takes its new money and buys guns, butter and builds bridges (to nowhere?). The money it spends winds up in the hands of people and businesses that in turn deposit it in the banking system. At this point the banks now have the money in their vaults and must decide what to do with it: make risky loans or hold riskless assets? In the past the only interest-paying riskless assets banks could hold were Treasury securities, but since 2008 they have the choice to put their excess reserves on deposit at the Fed and still earn interest on them. In fact, the rate paid by the Fed on excess reserves, although low at only 0.25%, is still higher than the derisory few basis points earned on Treasury securities all the way out to one year on the yield curve.
From the point of view of a bank, what's the practical difference between buying Treasury debt and putting the money on deposit at the Fed? I think the answer is pretty clear: There is no difference. Both have the same sovereign risk characteristics and the same liquidity. If the Fed is offering 25bps and the Treasury 5bps, excess reserves at the Fed will increase, and that's exactly what we've seen. And overall, riskless assets in the banking system have grown by about $2tn since 2008, mainly in the form of excess reserves. The fact is that banks don't (yet) want to make new loans, since the value of real estate collateral has continued to be weak, as has demand for credit from creditworthy borrowers.
So although there's no practical difference from the bank's point of view whether it holds Treasuries or reserve deposits at the Fed, it does have vastly different implications for the measurement of the monetary base: In the latter case the monetary base grows and in the former it doesn't.
At the moment we have the Treasury borrowing money from the Fed, which is in turn funded by the banks, who are ultimately funded by depositors, whose deposit growth is a result of debt-financed government spending. The Fed is nothing but a conduit through which the banking system is lending to the government. Banks have demonstrated that they don't want to make new loans to the private sector but instead want to increase their holdings of riskless assets. Importantly, this would be the case whether or not the Fed embarked on its policy of QE. Thus, if the Fed decided to unwind its QE the government debt that is currently sitting on the Fed's balance sheet is likely to end up on the balance sheet of the banking system, assuming that banks don't yet want to expand their loan books. The counterintuitive conclusion of this analysis is that QE doesn't matter!
Then where does inflation come from in this system? There are two channels. First, notice that on bank balance sheets, although private sector loans are still shrinking, loans to the government, either directly in the form of Treasury securities, or indirectly as excess reserves held at the Fed (and then lent on to the Treasury by the Fed), have grown more than private sector loans have shrunk. As a result, bank balance sheets in the aggregate have continued to grow throughout the post-crisis period. All this has been funded by healthy deposit growth, including both checking and savings deposits, so broader measures of the money supply are also growing. As noted above, this deposit growth is a direct result of increased spending by government.
Our fractional reserve banking system is replacing the private debt on its balance sheet with public debt. If the government continues to borrow and the banking system is willing to lend to it, this cycle has the potential to be highly inflationary: Since the capital requirements for holding Treasuries are effectively zero, the money multiplier on this additional borrowing is theoretically limited only by the desire of government to spend beyond its means and the willingness of banks to hold government debt.
The other channel by which inflation has percolated through the system is not directly related to the size of the Fed's or the banking system's balance sheets. Instead, overnight interest rates of close to zero have driven required returns on alternative financial and hard assets towards zero. I think we could argue about whether or not the Fed is artificially suppressing the level of interest rates---after all, banks seem more than happy to be rapidly expanding their loans to the government at interest rates of close to zero. Nonetheless, the current level of rates is undoubtedly leading to increases in the prices of certain scarce assets as the opportunity cost of holding these assets has disappeared. This perceived inflation is really a change in relative prices rather than a generalized inflation, though from the point of view of consumers of scarce commodities it probably feels about the same. It's important to understand that this commodity hoarding has nothing to do with QE, per se, but follows logically from the current structure of interest rates, which is not directly impacted by QE2 except insofar as the Fed is willing to hold longer term government debt than would be preferred by the banking system directly. Presumably an increase in the level of rates has the potential to deflate this bubble as rapidly as it inflated it.
The bottom line is that QE per se is not the source of inflation in this system. Inflation is the result of debt financed spending by the government accomodated and multiplied by the fractional reserve banking system, and by zero interest rates and resultant commodity hoarding. The end of QE in a couple of months is thus not by itself going to lead to a reversal in the rise of commodity and other asset prices that has characterized the past couple of years. Logically what we should be watching out for are signs of lower government borrowing and an explicit rise in interest rates. Might we be closer to one or both of those than the markets currently expect?
No comments:
Post a Comment