Monday, March 16, 2009

Absolute Confidence?

Not just the Chinese government, but every investor can have absolute confidence in the soundness of investments in the U.S.
– President Barack Obama, March 14, 2009, as reported in Bloomberg.

Statements like this one are causing me to lose confidence in the Obama administration’s economic policies. The particular investments about which the Chinese have been concerned are US Treasury securities. Absolute confidence in US Treasury securities is exactly what we don’t need. Absolute confidence in these securities is precisely the problem. The problem – for the US, anyhow – is that everyone wants to hold US Treasury securities instead of investing their money in productive activities (or spending it on the output of productive activities).

This is not just true of Americans; it is true of the world, including the Chinese. China has four choices:
  1. It can buy US Treasury securities.

  2. It can buy other US securities that represent productive uses of money.

  3. It can buy non-US securities, in which case the value of the dollar will fall and make US products more attractive, thereby encouraging Americans to invest in productive activities.

  4. Or it can buy no securities at all, in which case the value of the Yuan will rise, making non-Chinese products more attractive, thereby encouraging non-Chinese (including Americans) to invest in productive activities to replace the Chinese products that have become more expensive.
China has been choosing the first of these four options, and if it has “absolute confidence in the soundness of investments in the US,” then it will continue to choose that option, and Americans will continue not to invest in productive activities.

There is a common but misguided belief – to which President Obama, as one may surmise from his statement above, apparently subscribes – that a loss of confidence in US assets would have disastrous consequences for the US economy. In a boom time, or an inflationary time, that would be the case, but in a deflationary environment like the present, the consequences are more likely to be good. The Wall Street Journal, as an example, gives a typical statement: of the “loss of confidence would be a disaster” point of view:
In the worst-case scenario, a significant new aversion to U.S. investments could drive down the dollar and drive up interest rates, worsening the U.S. recession.
First of all, driving down the dollar would not worsen the recession; the direct effect would be to mitigate the recession by making US products more attractive. As for rising interest rates during a recession, that would indeed be a worst-case scenario, but it would not be the result of the aversion to US assets. It would be a result of a bad US policy response to that aversion.

My logic should be fairly clear:
  1. The Treasury has a choice whether to finance long-term or short-term

  2. The Fed has a policy – until further notice – of holding the federal funds rate below 0.25%, which policy requires it to purchase enough T-bills to keep short-term US Treasury rates near zero.

  3. Therefore, there is no limit on the Treasury’s choice of financing. It can issue as many or as few long-term securities as it chooses. What it does not finance long term, it will be able to finance short term at a near-zero interest rate, since the Fed will purchase enough T-bills to assure this.

  4. Therefore, the Treasury controls the supply of long-term Treasury securities.

  5. Therefore (assuming that the demand curve for such securities has the usual downward slope in the relevant range), the Treasury controls the price of long-term Treasury securities.

  6. Therefore (since bond yields – i.e., interest rates – depend inversely on prices), the Treasury controls the interest rates on its long-term securities.

  7. During a time of potentially deflationary recession, it will not be in the nation’s interest for the Treasury to allow interest rates on its long-term securities to rise, nor will it be in the nation’s interest for the Fed to allow short-term rates to rise.

  8. If they do so – whether or not they do so in response to a drop in demand for those securities – it is simply bad policy. Bad policy is not the result of declining confidence in US securities; it is the result of bad choices by policymakers.
So which interest rates are we talking about? Short-term Treasury interest rates? Those are controlled by the Fed and will not rise unless the Fed allows them to rise. Long-term Treasury interest rates? Those are controlled by the Treasury and will not rise unless the Treasury allows them to rise.

Or are we talking about private sector interest rates? Corporate bonds, as an example, are priced according to risk spreads over Treasury bonds. Those risk spreads depend on the amount of additional risk involved in owning corporate bonds and the amount of compensation that investors require for accepting that additional risk. The key word here is “additional.” A loss of confidence in US assets would most likely make US assets in general more risky. But how would it increase the additional risk of corporate bonds relative to government bonds? If anything it would do the opposite: the loss of confidence would weaken the dollar, making it easier for US corporations to sell their products, thereby increasing their profitability and their creditworthiness and reducing the additional risk from owning their bonds.

So – subject to exogenous changes in risk and in the price of risk – private sector interest rates will not rise either, unless policymakers allow them to rise. The only good reason to allow rates to rise would be if excess demand begins to lift the US out of its deflationary recession and to threaten it with excessive inflation – a scenario inconsistent with “worsening the US recession.” Loss of confidence in US assets is not the worst-case scenario; bad policy is. Under current circumstances, encouraging excessive confidence in US Treasury securities is itself an example of bad policy. It’s not the worst case, but it’s far from the best.



DISCLOSURE: Through my investment and management role in a Treasury directional pooled investment vehicle and through my role as Chief Economist at Atlantic Asset Management, which generally manages fixed income portfolios for its clients, I have direct or indirect interests in various fixed income instruments, which may be impacted by the issues discussed herein. The views expressed herein are entirely my own opinions and may not represent the views of Atlantic Asset Management.

Tuesday, March 10, 2009

Targets vs. Projections

There is a widespread view that the Fed’s “longer run projections” for the inflation rate can be interpreted as targets that the Fed will attempt to hit. The logic goes something like this. Suppose (as we shall presume) that the Fed has some target for the inflation rate but that it does not announce that target explicitly. The Fed will do its best to hit that target. It may not hit the target exactly: it may undershoot the target, or it may overshoot the target. Since the Fed is aiming directly for the target, the Fed is equally likely to undershoot the target by any given amount as to overshoot the target by the same amount. Therefore the target itself is also the Fed’s best “average” guess as to what the actual inflation rate will be. Thus, if the Fed makes a forecast (or a “projection”), we can conclude that the forecast is equal to the target.

Unfortunately, there is a flaw in this logic. The fact that the Fed is aiming directly for the target does not imply that the Fed is equally likely to undershoot as to overshoot the target by any given amount. If you’re driving directly down the middle of a lane but the right side of the lane is more slippery than the left, you’re more likely to skid to the right than to the left. From the Fed’s point of view, the possibility of undershooting its target should be considered more “slippery” than the possibility of overshooting the target.

If the Fed overshoots its target, it can tighten policy and push the inflation rate back toward its target, just as, if you start to veer to the left, you can turn the steering wheel to the right and get back in your lane. If the Fed undershoots its target, it may find itself in the same sort of liquidity trap that it is in today. In that case, policy may become ineffective, and the Fed may not be able to correct the undershoot. If you skid to the right, where the road is icy, then turning the steering wheel to the left immediately will not help you get back in your lane. So while your “target” is the middle of the lane, an “average forecast” would have to account for the fact that you’re more likely to miss that target on the right than on the left. Similarly, the Fed is more likely to undershoot its target than to overshoot.

So a target and a forecast are not the same thing. If the Fed were to release both a set of targets and a set of forecasts for future inflation rates, the targets should be higher than the forecasts. And if (as it has in fact done) the Fed releases only forecasts (or projections) and not targets, then, if we are to take the Fed at its word, and if the Fed agrees with the logic of my last paragraph, then we should conclude that its implicit inflation targets are higher than the inflation rates that appear in its projections.

Unfortunately, even if the Fed does agree with my logic, I don’t think we can take the Fed at its word. My impression is that the Fed is playing a language game in which all parties have implicitly agreed that the word “projection” will be used to mean “target.”

If this is true, then it’s bad news, because it means that the average expected inflation rate over the “longer run” will be less than the Fed’s projected “central tendency” of 1.7% to 2%. And more specifically, it means that the risk of deflation will never be entirely gone. If you’re on a four-lane highway and the right shoulder is icy, you’re better off driving in the left lane, where there is minimal risk of veering onto the icy part. But the Fed has declared its intention to keep driving in the right lane – at less than 2% inflation, right next to that icy place where a severe recession (much like the one we are currently experiencing) could render policy ineffective.


DISCLOSURE: Through my investment and management role in a Treasury directional pooled investment vehicle and through my role as Chief Economist at Atlantic Asset Management, which generally manages fixed income portfolios for its clients, I have direct or indirect interests in various fixed income instruments, which may be impacted by the issues discussed herein. The views expressed herein are entirely my own opinions and may not represent the views of Atlantic Asset Management.