source: www.hussmanfunds.com
As far as it goes there isn't much to argue with here. Clearly there is a strong inverse relationship between the demand for money and the level of interest rates---as rates fall the opportunity cost of holding money falls and the impetus to put one's money to work recedes. Using this as a starting point, let's think through the implications of a potential rise in rates, now that the size of the Fed's balance sheet (i.e., the monetary base) has expanded so dramatically in the recent past. Given that we're currently in the lower right corner of this graph, even a modest increase in rates implies a sharp rise in nominal GDP absent a significant reduction in the monetary base---in other words, we should expect a big rise in prices.
Now let's think back to the previous post and ask why this might not be the case today. How is the level of reserves determined? Well, the simple answer is that the Fed creates them in the process of buying Treasuries. In fact, it can be shown that the level of reserves in the banking system as a whole is almost completely determined by the central bank's decision to sell or purchase assets (usually Treasuries but in recent years including various other assets). Although it's possible for an individual bank to reduce its reserves, it's not possible for the banking system in aggregate. There's a good explanation of this in a July, 2009 paper from the New York Fed.
But, as discussed last time, from the perspective of the banking system, in a world where reserves earn interest they are also a near substitute for Treasuries. Banks decide on the optimal mix of risky and riskless assets. In this framework the level of reserves, which we know is determined by central bank policy, is simply a component of the banking system's riskless assets. Since today banks are continuing to increase their holdings of Treasuries, over and above the buildup of their reserves resulting from the Fed's policy actions, we can assume that the banking system wants to hold at least as many riskless assets as the Fed has created in reserves. Thus, if the Fed hadn't bought all those Treasury securities, the banking system would have done it instead.
The above logic is why I feel confident in saying that, given the current circumstances, QE hasn't mattered---the Fed has just bought riskless assets that the banking system demanded anyway. Of course, one could easily imagine a scenario wherein the Fed is growing its assets faster than the banking system would choose to, leading to an effort by banks to reduce the relative size of their reserves by expanding their balance sheets. In fact, in looking at the state of the banking system today I wonder if we're not getting there. Absent a reduction in the size of the public sector borrowing requirement or a rise in interest rates that is likely to be in our not-so-distant future.
So now what could happen when interest rates rise? Should we be concerned about latent inflationary pressure, in the absence of a proactive reduction in the level of reserves? In a world where reserves earn the same rate of interest as Treasuries the answer is no, since a rise in interest rates has no bearing on a bank's willingness to hold reserves (the rate it earns on those reserves goes up). Of course, if the Fed ceased paying interest on reserves, or paid a rate much below the rate that banks could earn on other riskless assets, banks would attempt to reduce the relative size of their reserves, and the only way the system can do that in aggregate is by expanding its balance sheet. The result would undoubtedly be inflationary.
All of this is a long-winded way of saying that as long as bank reserves earn interest comparable to Treasuries, a rise in interest rates is unlikely to trigger inflation, contrary to what is predicted by economic theory. How's that for a mind twister?
No comments:
Post a Comment