Bloomberg has a story that asserts that the Fed, by taking some of the “froth” (Williams) out of the bond market, has bought space to continue with QE. Surely, the Fed doesn’t buy the bond-bubble argument? Surely, at least, Bernanke doesn’t?

I have no idea what the Fed thinks it is doing. I have no idea why it started QE3 — I was confused at the time, and I remain so.

To be fair to the Fed, let’s think about why you might do QE.

First, you might do it because you have a portfolio-balance theory of bond prices: in other words, for the technical effect that your buying will have on the market price. But this is simplistic. The fact that the Fed is buying bonds shows that it is in loosening mode, which ought to boost economic expectations, and therefore bond yields. That is exactly what happened, for example, in H1 2009, and in late 2010-Q1 2011 during QE2. The idea that QE should reduce bond yields, relative to what would have happened otherwise, looks pretty shaky to me.

Second, you might do it because it is a signal of your intention to keep short-term rates low, in order to flatten the yield curve. This seems to have been at least part of the rationale for the BoJ’s QE in the 2000s. But it also doesn’t seem to work — because long-term rates rose in the previous QE episodes, the yield curve steepened.

Third, you might do it in order to raise inflation expectations (either by signalling something about your reaction function, or by using the false opinions of the inflationistas to your advantage). Why would you want to do that? In order to signal your commitment to avoiding deflation, or to keep inflation expectations up in order to keep real interest rates down. The first two episodes of QE could fit with the latter rationale — in 2009, the 10-year breakeven was negative and the real 10-year interest rate was above 1.5%, and in 2010 the breakeven dropped sharply to 1.5%, with lower values at the shorter end. But at the institution of QE3, the 10-year breakeven was over 2% and the 10-year real rate was negative and apparently falling (because, I think, of forward guidance, which is more effective than QE in this regard).

Fourth, you might do it in order to raise asset prices through a “money flow” effect. I am sceptical of this idea. We can see the idea in balance sheets as follows:

Assets: +bonds
Liabs: +reserves

Assets: +reserves
Liabs: +deposits

Assets: -bonds, +deposits

The idea is presumably that the extra deposits should be spent, on investment assets or houses. But a) as my blog post earlier today shows, there is no obvious relationship between the equity market and money-supply growth, and b) it is not necessary for households to spend their cash at all: it is not much different from bonds at present, but has no duration, so the sale of bonds at extremely low interest rates could simply be a portfolio reallocation.

Perhaps there ought to be an important money-flow effect into the mortgage market from MBS purchases. But, according to the MBA, mortgage refinancing applications were already relatively high in late 2012, having been so all through the year. The mortgage purchase index did increase a little from its lacklustre level from the end of 2012, but hardly dramatically.

I could have this all wrong, but I just cannot get my head around why money flow ought to be important for the market. This fits with my general presumption against technical explanations for market trends: I think that changes in collective judgements of value are generally far more important than technical factors in causing large changes in market prices.

So, since there was no clear point to QE3, there is no clear reason for it to end. At the time, the talk was all about the weakness of the labour market, so perhaps the Fed hoped the programme would affect that market via some mysterious set of mechanisms. But even that is not consistent with the move towards tapering now. Improvement in the prospects for the labour market was a condition for the end of the programme, and it is not clear that the prospects have improved very much.

As a result, the Fed’s reaction function, as Tim Duy says, has become unclear. It appears to include a (vague) financial-stability factor, or perhaps, as Isabella Kaminska suggests, a market-liquidity factor. This is presently the cause of much confusion.

The most reasonable interpretation seems to me to be that the Fed felt a political imperative to “do something” about the continued high level of unemployment, and fell into the Politician’s Fallacy (something must be done; this is something; therefore, we must do it). But, as with any person or institution that does something without a clear reason, it has thought of a whole lot of reasons to fret about its decision, and wants to bring the programme to a close. Its problem, however, is that it cannot do that without asserting that the outlook for the labour market has really improved, and thereby raising expectations for the first hike in interest rates while at the same time looking relatively more hawkish, and therefore reducing inflation expectations. The net result is a jump in real interest rates, and a significant de-facto tightening of monetary policy (see the 5-year real interest rate, here: In other words, the start of QE3 was a policy mistake, but its end is a major policy mistake that could have negative implications for the US economy in the coming months. Combined with the fiscal tightening, whose effects may be delayed, it seems to me that monetary and fiscal policy could lead to a much weaker economy than the markets are currently pricing.

I could have much more to say today, but this can’t go on for ever! So just an interesting observation. FT Alphaville reports a chart from Deutsche Bank showing a spike in short-term portfolio flows into the US in recent times (the scale is not very clear: This would cohere with my idea that the sharp rise in USD/JPY was the result, somehow, of events in Japan; that it indicated inflows into the US; and that those inflows were the reason for the strength of equity markets in H1 2013. I shorted equities at the recent top, because USD/JPY had fallen and Abe’s Third Arrow had disappointed, and equities duly fell. Now they have risen again, with USD/JPY. Perhaps I should be short again.

But, why, why why? Who is putting money into the US? Is it the Japanese? Is it carry traders? I can’t find evidence for any of this. Perhaps I should stop worrying and go with the theory: if USD/JPY stays lacklustre, short equities. But I do want more details!


  • Case-Shiller 12.2% d.e. but growth rate rose to a new post-crisis high.
  • CB consumer confidence fell slightly from its post-crisis high last month.
  • UK GfK consumer confidence continued its sharp improvement, which has now gone on for three months.
  • German unemployment fell more than expected, Jun.
  • Eurozone CPI flash 1.6% YOY a.e. Jul.
  • Eurozone unemployment was flat at 12.1% b.e. Jul.