Friday, February 13, 2009

Price Level Targeting: An Example

In my previous post, I tried to make a case for using price level targets as a way to overcome the problem of money hoarding (which, under present circumstances, should be understood to include T-bill hoarding, and perhaps Treasury security hoarding in general). In this post I’m going to give an example of what a set of price targets might be and how it would work. You might even consider this a recommendation – although I wouldn’t recommend myself as the best person to recommend a specific set of price level targets.

In general, there are four characteristics that a good set of price level targets should have:
  1. It should be realistic. We might love to see a 3 percent inflation rate for 2009, but that’s not going to happen, no matter what the Fed does. To be credible, the targets must allow for a couple of years of low inflation, simply because we know that the Fed has little influence over the inflation rate in the short run. In general, the targets should be reasonably conservative relative to what might be possible, because, if the early targets are not hit, the Fed will have more difficulty hitting the later targets. We don’t want to set the Fed an impossible task in the case where things go badly at first.

  2. It should target prices high enough to make people nervous, even if they don’t think the Fed can hit those targets. This is where my poker game analogy (as discussed in the previous post) comes in. You may think that the player with the four hearts up is probably bluffing, but if she throws enough blue chips into the pot, and if you’re a conservative player, you’re going to fold your three aces anyway. These days almost everyone is a conservative player. So if the Fed sets the price level target high enough, just the risk that it might hit that target should be enough to motivate investors to hold nonmonetary assets. The choice today is between risky assets with a high expected return (stocks, high-yield bonds, etc.) and safe assets with a low expected return (cash, Treasuries, etc.). Set a price level target high enough, and the choice – even if you don’t have much confidence in the Fed’s ability to hit its targets – will be between two risky assets, one with a high expected return and the other with a low expected return.

  3. It should transition to a “normal” rate of inflation in the long run. A normal rate of inflation used to be 2 percent, but recent experience shows that 2 percent is dangerously close to zero. I think 3 percent is the lowest reasonable target for a long-run inflation rate, and that’s what I will assume in my example, because it’s close to what most people would think of as normal. If it were up to me, I would choose a long-run inflation target of 4 or 5 percent. (I’m using the phrase “inflation target” loosely. I mean the inflation rate implied by long-horizon price-level targets.) It’s an open question how long exactly the long run is. In my example, I’m choosing a set of targets that implies a 3 percent inflation rate starting in year 14 (2022, since 2009 is year 1). Note that the 3 percent implication is conditional on earlier targets being met. If the Fed falls short of the price target for the end of 2021, the inflation rate will have to remain higher than 3 percent for a while thereafter in order to catch up.

  4. It should target prices low enough that they don’t imply ridiculously high inflation rates. After a while it should be technically feasible for the Fed to engineer an inflation rate as high as, say, 20 percent, and if that were expected to happen, it would certainly discourage people from holding monetary assets. But at some point the cure becomes worse than the disease. If we’re targeting a long-run implied inflation rate of 3 percent, the transition from 20 percent to 3 percent would likely be very unpleasant. And at some point also, it becomes too unfair to ask today’s long-horizon creditors and fixed income recipients to bear such a large part of the burden for fixing the economy. The targets in my example imply a maximum annual inflation rate of 10 percent (conditional on earlier targets being met), a maximum (conditional) 5-year average inflation rate of 8.4 percent, and a maximum (conditional) 10-year average inflation rate of 6.7 percent. I would say that those are high (as point 2 requires) but not ridiculously high. (Others may have a different opinion.)
The targets in my example are for the core CPI (Consumer Price Index excluding food and energy), and there is a target for each December over the next 20 years. The targets may be understood relative to the actual reported core CPI (216.1) for December 2008. The table below shows the targets and their implications:

ABCDEFG
priceimpliedimpliedimpliedimpliedimplied
targetannualaverageinflationinflationaverage
coreinflationinflationtargettargetinflation
CPI)targettargetoverovertarget
frompriorpriorfrom
year zero5 years10 yearsyear 5
assuming
2008216.1zero
2009219.31.5%1.5%inflation
2010223.72.0%1.7%
from
2011230.43.0%2.2%year 0
2012239.74.0%2.6%to year 5
2013254.06.0%3.3%3.3%
2014274.48.0%4.1%4.6%27.0%
2015301.810.0%4.9%6.2%18.2%
2016329.09.0%5.4%7.4%15.0%
2017355.38.0%5.7%8.2%13.2%
2018380.17.0%5.8%8.4%5.8%12.0%
2019403.06.0%5.8%8.0%6.3%
10.9%
2020423.15.0%5.8%7.0%6.6%10.1%
2021
440.04.0%5.6%6.0%6.7%9.3%
2022453.23.0%
5.4%5.0%6.6%8.6%
2023466.83.0%5.3%4.2%6.3%8.0%
2024480.83.0%5.1%3.6%5.8%7.5%
2025495.33.0%5.0%3.2%5.1%7.2%
2026510.13.0%4.9%
3.0%4.5%6.8%
2027525.43.0%4.8%3.0%4.0%6.6%
2028541.23.0%4.7%3.0%3.6%6.3%

These targets imply a 5.8 percent average annual (compound) inflation rate over the next 10 years. (See the entry for December 2018 in column D.) Suppose the Fed were also to target the 10-year Treasury yield at zero (a target the Fed could certainly achieve by intervening directly in the longer term Treasury market, although meeting that target could conceivably require the Fed to buy up almost all of the national debt). The result, if the Fed were to hit its price level targets, would be a real (inflation adjusted) 10-year Treasury return of -5.8 percent.

Markets would not have full confidence in the Fed’s ability to hit its price targets, but the threat of a -5.8 percent real return, even with only limited credibility, should be enough to make people nervous. And the more nervous people get about holding monetary assets (in which term I mean to include Treasury notes), the more they will invest in real assets (or just spend money), and the easier it will be for the Fed to hit its target. It might not be necessary to set the target real return as low as -5.8 percent (i.e. to target the 10-year Treasury yield at zero), but the price level targets would give the Fed room to maneuver.

To get an idea of how the “catch up” process would work with these targets, consider the case where the average inflation rate over the first 5 years is zero. In December 2013, the Fed will miss the target by a factor of 0.85 (the actual core CPI of 216.1 divided by the target of 254.0). The Fed will then have to make to make up that 15 percent “deficit” over the subsequent years by pushing the inflation rate above the original “planned for” inflation rates. Column G shows the average annual inflation rates that would be necessary to make up the deficit in a given number of years. To make it up in one year (obviously unrealistic) would require an inflation rate of 27 percent. To make it up in five years (fairly realistic) would require an inflation rate of 12 percent.

To get an idea of the overall pattern of inflation rates implied by this set of targets, we can divide the next 20 years into 5-year intervals and look at what the average inflation rate would be in each of those intervals if all targets are hit:
  • Dec-2008 to Dec2013   3.3%
  • Dec-2003 to Dec2018   8.4%
  • Dec-2018 to Dec2023   4.2%
  • Dec-2023 to Dec2028   3.0%
So the major “push” comes in the second five-year period. The inflation rate over the first 5 years has to be relatively low just because the inflation rate is currently low, and it’s difficult or impossible to increase the inflation rate very quickly. Over the following 5 years, the inflation rate rises to its peak of 10 percent in 2015 and then gradually comes down thereafter.

The hope is that the transition from high to normal inflation (after 2015) will not be too difficult, because it will be fully anticipated. Individual businesses will expect a slowdown in the overall inflation rate and will therefore slow down their own price and wage growth, allowing the anticipated monetary policy to support a normal level of output and employment. That’s the theory, anyhow.




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.

270 comments:

«Oldest   ‹Older   401 – 270 of 270
«Oldest ‹Older   401 – 270 of 270   Newer› Newest»