Saturday, July 19, 2008

Too big to fail

Besides the extraordinary rises in commodity prices and near-tripling of the U.S. federal government's default risk, another important factor for those interested the world's economy to keep in mind is that foreign governments have been bailing out the U.S. federal government by buying more of its Treasury bonds. Much of the increase in inflation expectations, and now in overt default expectations, reflected in commodity prices probably comes from the possibility that this ongoing foreign bailout of the U.S. may at some point stop and reverse. If foreign central banks and sovereign funds suddendly dumped their Treasuries -- which paying a negative real interest rate and exposing one to dollar risk are after all a rather bad investment -- either the Federal Reserve would have to madly "print" trillions of dollars to buy up these Treasuries, causing oil prices to skyrocket still further and retail gasoline prices to quite possibly jump each week by dollars per gallon, until American cars, trucks, and airplanes would be useless, or the federal government, facing extraordinary costs of funding its deficits, would have to overtly default. The problem the Federal Reserve faces is that with modern digital markets any inflationary behavior on its part is reflected in oil prices as soon as any big money learns about said behavior, and is thus usually reflected at the retail gas pump within a week. I wrote in the above-linked post about sovereign debt risk of "covert default" through inflation, but that is an obsolete phrase -- you can go down to the local gas station and see these de facto federal defaults in progress. If foreign governments thought the Federal Reserve was going to hyperinflate, that would probably cause them to dump the Treasuries in a kind of self-fulfilling prophecy. It would be a run on the federal government itself, the biggest run on a bank the world has ever seen. This is not a probable risk, but it has now become a quite signficant risk.

The ongoing foreign bailout of the U.S. government by massive purchases of Treasury debt helps explain two things: (1) why U.S. Treasuries and bank rates related to them don't seem to be reflecting inflation expectations (i.e. why they are paying negative real interest rates compared to the expected inflation suggested by commodity prices) -- their price is being held artificially low, not just by Federal Reserve net purchases but by foreign central bank net and sovereign fund purchases as well. Add to this the natural reduction in their interest rate due to the credit crunch causing a flight for safety (in a credit cruch people are willing to pay even negative real interest rates for a relatively safe investment). The foreign bailout also explains (2) why commodity prices have raced so far ahead of PPI and CPI inflation. Commodity prices may reflect, not so much increases in the dollar supply beyond dollar demand that have already happened, nor even expectations of a steady future consumer inflation that will happen eventually, nor an iminent severe consumer inflation is sure to happen soon, but expectations of a severe consumer inflation that mostly hasn't happened yet and may not happen, and of a possible Zimbabwe/Weimar-style hyperinflation that probably won't happen. But the increase in the probability of severe inflation combined with a great increase in the formerly extremely small (and still small, but not extremely so) probabilities of the more extreme hyperinflation still averages out to rather high inflation expectations.

Why are central bankers bailing out the U.S. federal government and its dollar rather than acting like rational profit maximizers? Central bankers and government treasury officials are very political animals, and they fear the disaster they believe would ensue should the "full faith and credit of the United States [federal government]" fail. It is the most unthinkable economic event for the entity with the most collosal budget, and the dominant threatener and wielder of military force in the world, to go under. The U.S. federal government has now deemed Bear Stearns, Fannie Mae, and Freddie Mac "too big to fail" and has bailed them out. That federal government itself is the ultimate entity that is "too big to fail", and thus foreign governments and central bankers are bailing it out, and will probably (but hardly surely) continue to do.

Extrapolating from oil prices rises (i.e. assuming the entire oil price rise has been due to monetary factors rather than to consumption, production cost, or non-monetary political fundamentals) using the net present value formula over 50 years, I estimate that long-run dollar inflation expectations have risen about 5%/year from 1998, when there the Asian crisis caused a flight to the dollar and many people thought deflation was the bigger worry, to today. This might for example reflect an increase in expectations from 0% to 5%/year, or 1% to 6%/year, etc. -- I don't have a way to estimate the absolute value. Gold price rises over the last decade give a slightly smaller rise in inflation expectations, of about 4%/year. Note that this doesn't mean 5% or 6%/year CPI or PPI inflation over the next 50 years is inevitable -- increased expectations reflect a range of probabilities, especially very heightened increases in probability estimates of extreme (>10%/year) and hyper (>100%/year) inflation in the dollar. It's also still significantly possible that the ultimate lagging indicator, "core inflation", will stay in the 2-3%/year range and the whole storm will blow over.

Besides the great rise in commodity prices, in particular the most "money-like" commodities oil and precious metals, the foreign bailout of the U.S. federal government is another reason to believe that all major currencies are falling against a hypothetical stable standard of value, not just the dollar. As foreign central banks buy more U.S. Treasuries they take on more of the risk associated with the dollar, so we should expect the floating currencies these central banks issue to move more in correlation with each other. Indeed, the dollar has held its own against the euro and against on average other major currencies since February as foreign central bank purchases of Treasuries have accelerated.

People interested in my writings about money might also want to check out my classic essay on monetary origins, "Shelling Out: The Origins of Money", my classic essay on micropayments, and my overview of the monetary reasons for the broad-based commodity price increases over the last decade and in particular over the last five years.


  1. Anonymous8:58 PM

    I've wondered if the recent run up in oil prices, represents not a sudden surge in money creation, but rather asset relocation after the real estate collapse. People used to believe that investing in stocks and real estate provided good hedges against inflation. But the trouble is, those sectors can get overproduced, and prices eventually fall to marginal cost of production. Thus investors now are reallocating towards commodities, because those do not get overproduced.

    So it's not that inflation is any different than five or fifteen years ago. It's just that now people are investing in commodities as inflation hedges, rather than stocks and real estate. That explains why stocks, real estate, and other inflation indicators are all down. I would not be surprised if the money supply is actually deflating as the credit crunch has slowed the amount of new money creation. Perhaps now is the time to put money back into safer bonds.

  2. Anonymous3:36 PM

    libra: People used to believe that investing in stocks and real estate provided good hedges against inflation. But the trouble is, those sectors can get overproduced, and prices eventually fall to marginal cost of production. Thus investors now are reallocating towards commodities, because those do not get overproduced.

    I agree that to some extent this is probably true. This is reinforced by the fact that commodities are a smaller market than real estate or stocks, so only a relatively small move out of them is needed to produce a relatively big run up in commodities. A caveat is that (speaking very generally) minerals will also be overproduced if their prices exceed the net present value of the mineral in the ground given inflation expectations (see my charts in my subsequent blog post).

    So it's not that inflation is any different than five or fifteen years ago. It's just that now people are investing in commodities as inflation hedges, rather than stocks and real estate.

    If you mean that rational inflation expectations are no higher, because stocks and real estate are down, then commodity prices exceed their net present value given inflation expectations and we are in a commodities bubble. The fact that minerals production has not substantially increased in response to higher prices tells me that it's not a bubble, i.e. that the the prices reflect rationally higher inflation expectations. These inflation expectations can be rationally explained by the extreme potentials now possible in federal finance that I've outlined above.

    But in the U.S. at least real estate is not down at least nominally over its prices 3 years ago and longer, it is primarily down compared a year ago. The real estate market had been pumped full of moral hazard that it is now disgorging before it will go back to being as reliable an inflation hedge as it once was.

    As for stocks, I've never thought they were a good inflation hedge. Price stickiness means they can only raise their prices far more slowly than the costs of their inputs, especially mineral inputs, so inflation (or even just higher inflation expectations which drive up commodity prices first) gobbles up profit margins. In the 1970s, U.S. stocks lost nearly 90% of their real value, as much as in the Great Depression, but unlike the Great Depression they broke even in nominal value. Inflation can destroy real stock values just as much as deflation.

  3. Anonymous3:52 PM

    hi nick
    just thought you might want to know
    "Shelling Out" link is not working.

    i have a link reference to it on my site, so i am keen to refer to another link if the essay is on the net