Wednesday, September 21, 2011

When bad men combine, the good must associate

It seems as though good men today are silent, weak and divided, while evil men are full of sound and fury. Rick Perry is the only important politician today who understands macroeconomics. Before you laugh, hear me out:

1) Governor Perry believes that the Fed can create as much inflation as it wants to, whether or not we are at the "zero bound"
2) Governor Perry believes that this inflation will have real consequences
3) Governor Perry believes that the real consequences of increasing inflation will be to increase output and employment

The only difference between us is that I think these (particularly the last) are good things. Governor Perry is trying to unseat an incumbent president and doesn't give a damn about the public good, therefore he has warned/threatened the Fed to leave things as they are and to continue strangling the economy with tight money.

Meanwhile, Mitt Romney and the Democrats are bumbling idiots who think money is easy because interest rates are low. I may be wrong about Romney; he may simply be holding his tongue for fear of angering the Republican mob. There is no explanation for the silence of Democrats on monetary policy other than pure, unfettered stupidity however. They are 12 months away from losing the presidency due to poor economic performance, and anybody with eyes can see what the problem is. First up, the all-too-familiar TIPS spread timeseries along with the SPX (edit: please see note at the end of the post for data sources):

Okay,  straightforward enough. The general story is clear: when monetary policy is functioning well, equities (as a proxy for expected future NGDP) aren't particularly sensitive to expected inflation. "Something bad happened" in late 2008, and expected inflation and NGDP dropped off a cliff. Since then, the Fed has run a tight money policy, constraining NGDP expectations, and turning equities markets into the correlation one monsters we see today. But anybody paying attention has already figured out this much, and as a quant I want more detail. 

If you want to see how bad things have gotten, you can use the beta of equities vs TIPS spreads. Loosely speaking, this beta is the amount the SPX is expected to increase for a 1% increase in TIPS spreads. If beta equals 10, then the Fed can increase equities prices by 10% by driving up the TIPS spread by 1%. Technically, I am computing the moving ratio of 6 month covariance(TIPS spread, SPX) to 6 month variance(TIPS spread) on a 1 week sampling frequency. 

Now, a well-functioning monetary policy shouldn't have a beta below 0 or above 5 against 5yr TIPS spreads (depending on your discounting any of the numbers in that range are "ideal", meaning that real variables will be independent of inflation expectations). Well, let's see what's actually happening:

Good God! The data's a bit noisy, but according to my read of this, the Fed is massively constraining real activity through its insane tight money policy. A 1% increase in inflation for 5 years (total increase in expected price level of 5%) is expected to lead to ~20% higher nominal output (integrated over time). That means that we get at least 3% real growth for 1% inflation. I want Obama to lose come Nov 2012 as much as anybody working on Wall Street because I think he's mucked around on the AS side of the equation enough to do real damage, but I'm not willing to cut off my nose to spite my face with more unnecessary AD-induced doldrums.

Note on data sources: for 5 year tips spread I used the data on For the SPX closes I used^GSPC+Historical+Prices. The TIPS spread overstates expected inflation, and does so more when inflation expectations are close to 0 because of the embedded optionality inside TIPS. However, I know of no publicly available data which would correct this bias (either by providing implied vol of inflation or by directly giving prices of delta one instruments like inflation swaps). 

Tuesday, November 30, 2010

A subject I am completely unqualified to comment on seriously

Instead of talking about something which would play to my core strengths (say, energy economics, financial regulation, complex derivatives, risk management) I'm going to take the opportunity to demonstrate my ignorance of macroeconomics in general and monetary policy in particular.

Now, the obvious impetus for this post is QE2. The Ben Bernank has decided to run the printing presses and buy Treasuries out to 10 years maturity to drive their yields down. This has created much wailing and gnashing of teeth amongst some inflation hawks.

In some respects, I understand how quantitative easing works a hell of a lot better than I understand how "regular" monetary policy works. In normal open-market operations the Fed messes with the extreme short end of the yield curve. They can fiddle with that as much as they want, but the long end of the curve barely moves. Why doesn't it move? Well, because fixed income traders aren't (complete) idiots. They know that the Fed is going to be there to take the punch bowl away as soon people start having too much fun. If Fed manipulation is expected to work in the short-to-medium term then forward rates are barely affected by the Fed's signaling. Only if Fed actions are expected to fail (necessitating low short rates for an extended period of time) will their actions have an actual effect on the interest rates businesses in the real economy care about. It's nice to be able to borrow cheaply for the next 3 months, but it's pretty hard to build factories, houses and other stuff on that sort of timescale. So there's some kind of black magic going on with normal monetary policy I don't understand. Maybe nobody understands it, or maybe I'm just obtuse. With QE, on the other hand, the Fed just steps in and miracles the yield curve into whatever shape it wants. Abra-damn-cadabra. That's the kind of power you wield when you have access to an unlimited supply of money.

Why does a change in the yield curve for government obligations have any effect on the "real economy"? Well, I know of a few different transmission mechanisms:

1) Cheaper borrowing: the most straightforward story. Holding credit risk (and the mumbo-jumbo that goes into the swap spread) constant, a change in gov't bond yields leads to an equal change in the borrowing costs for businesses. This encourages expansion, the creation of new businesses etc. An alternative means of looking at this is that the present value of future cashflows are discounted at lower rates, meaning that investment becomes more attractive (ceteris parebus). Basically, the Fed makes holding Treasuries less attractive relative to other investments. Money flows into investment and consumption. Inflation goes back up to "normal" levels. More stuff is bought and built. Everybody's happy.

2) Tobin's "q". Entry of the Fed's new money into the capital markets causes bond prices to go up. The extra money they dumped into treasuries has to go somewhere, and one of the places it goes is the stock market. Equity prices also go up. All of a sudden, companies can finance more investment by issuing new equity than they could have previously.

3) Wealth effects. A stock market and bond market where investors earn paper profits causes them to feel richer. What do they do? Spend more money, of course.

4) The Plausibly Deniable Mechanism: exchange rates. It doesn't work in the long term, but in the short term an unexpected shock in nominal interest rates causes people to dump their dollars, causing the price of USD-denominated goods to drop outside the USD zone and non-USD denominated goods to rise inside the USD zone. This causes an increase in demand in the USD zone. Why do people dump their dollars? Well, usually what I hear is some mealy-mouthed explanation about inflation expectations. The mechanism seems to be a lot clearer to me when you say two little words: "carry trade". You know the story about the Japanese housewife? She's the one who trades her Yen in for some Aussie dollars, puts those in an Aussie bank and earns 5% more than her Yen would have earned in a Japanese bank. (Side note: what's the risk here? Well, this isn't an arbitrage, because the JPY/AUD forward curve slopes upward at 5%. The danger is that JPY will actually appreciate along its forward curve, screwing all of those Japanese housewives). If the interest rate in dollars drops, people sell their dollars and buy AUD or any other currency with higher interest rates. Of course, Uncle Ben would never do anything this underhanded to the USA's nice trading partners, would he?

To summarize: QE is pretty straightforward, and in my opinion is probably desirable at the current moment in time (inflation is very low, output growth is mediocre, employment is weak). It's the regular open market operations of the Fed that are mysterious (to me, at least)...