Marginal Revolution

Department of Uh-Oh, The End of the Mortgage Agencies

Fannie and Freddie’s preferred shares have been considered so safe that banking regulators let banks count them in the capital required as a cushion against loan losses.

That should read "*had* been considered so safe."  Further background is here and here.  We also have an impossibility theorem.  I do not see how our government can let the value of this preferred stock fall much further, given the extent to which bank insolvency would increase.  I also do not see how our government can prop up the value of this preferred stock to a significant degree.  Silly me.  I guess that means I bet on the latter impossibility.

Here is yet further discussion of the end game.  It seems the common stock owners will end up suffering more dilution than they had been expecting.  The $340 billion in agency debt held by the Chinese central bank will be protected, as it must be.  Have a nice day.

And the newspapers were wondering why the Dow tanked 345 points.

Hail Giacomo Ponzetto!

He comments:

...Kotchen and Burger seem to ignore that the oil-drilling problem is a textbook application of dynamic programming. In particular, Dixit and Pindyck (1994, ch. 12) highlight the analogy between an undeveloped oil reserve and a call option on a dividend-paying stock. They conclude, referencing Paddock, Siegel and Smith's (1988) original study, that "the use of standard NPV methods would lead to substantial undervaluation of the reserve as well as premature development."

The rest is under the fold...

Since the option is perpetual, a closed-form solution is easy to obtain if one makes standard assumptions: production from a developed reserve is represented by exponential decline, the price of oil is a geometric Brownian motion, asset markets span, etc.

Following the authors cited above, assume:
-- a payout yield of 4% from a developed field;
-- a risk-free real interest rate of 1.25%;
-- a volatility of 0.2
Then the option value of waiting is such that we should only drill when the present value of the developed reserve is at least 1.6195 times the cost of developing it.

Suppose that the price of oil follows a martingale, so the current price of $105 per barrel is also the expected future price at any time. Suppose the ANWR reserve comprises 7.06bn barrels and that once the oilfield is developed it will pump out 5% of the reserve every year at a constant marginal cost of $5 per barrel

Then at the 1.25% discount rate the developed reserve is worth $564.8bn (which is reasonably close to Tyler's $600bn estimate). However, if we start drilling now the reserve will be developed in 10 years (EIA 2004, 2008), so we must calculate the present value of this sum. The correct discount rate here is the payout rate of 4% (Dixit and Pindyck 1994, p. 403), so the NPV of drilling now is $378.6bn. Therefore, the option value due to volatile oil prices implies that we should drill now only if the cost of drilling, including its environmental impact, is below $233.78bn.

Since the cost estimate above ($5 for getting a barrel of oil to market from an existing well in Alaska) only accounts for an NPV of $18.93bn, Kotchen and Burger's figures leave me with a $103.87bn cost of developing the reserve. Then we should drill now if the environmental cost is less then $130bn, or the willingness to accept compensation to allow drilling less than $590 per voter.

Admittedly all my figures are very rough estimates, but I believe this is the correct order of magnitude. The reserve is indeed worth about $600bn, but that is not very important, because the choice is not between drilling now or foregoing drilling forever, but between drilling now or waiting and seeing.

Furthermore I have ignored the possibility of cost-reducing technical progress. I don't see why drilling should become costlier or more environmentally damaging; but it probably could become more efficient on either account. That would increase the option value of waiting.

Obviously, we should rush to drill now if we expected oil prices to decline sharply in the future, because then the reserve would be rapidly depreciating while it is left in the ground. But that does not seem to be the argument of the bozos on either side of the aisle.

It's also worth noting:

1. Critics of drilling usually want to shut down the option forever and the political window cannot be expected to remain open forever.

2. There is a general global warming case against developing the resource.  Note that supply restrictions can be far more effective than a Pigou tax.  A Pigou tax doesn't guarantee the stuff won't be pumped anyway, albeit at lower profit.

3. The Pigouvian case against developing ANWR makes sense only if we are taking other systematic actions to raise the price of fossil fuels and restrict fossil fuel use.  Otherwise we may just be leaving a $600 billion dollar bill on the proverbial sidewalk.  This may be a classic case of twin-peaked preferences.

4. Depending how the money is spent, and on the general equilibrium properties of the system, it still may make sense to have a) a Pigou tax on fossil fuels, and b) ANWR development.  For one thing, it does matter who captures the profits from fossil fuel development.  You could imagine an even stiffer tax on imported fossil fuels (relative to what would be optimal without ANWR), combined with ANWR development.  You can spin out lots of tricky problems here.

Magnus #1?

17 year old Magnus Carlsen won a chess game today and is probably now, unofficially, ranked #1 in the world.  (World champion Anand lost and fell behind in rating points.)  Here is an illuminating recent profile of Magnus.  I believe Paul Samuelson is the closest to an economics prodigy we have had.  He was thirty-two when his Foundations of Economic Analysis was published but I have heard that he wrote the book at a much earlier age (does anyone know the exact age?).  He was probably one of the best economists in the world when he received his undergraduate degree at Chicago at the age of 20.  Frank Ramsey is another example of an economics prodigy although he didn't even think of himself as an economist per se.  Can you think of other prodigies in mathematical economics?  I attribute their scarcity to the relative aesthetic poverty of mathematical economics (for most people it's not that fun or beautiful) rather than the need for complementary experience-acquired wisdom.  Do you agree?

Addendum: Andrew Gelman considers statistics.

Assorted links

1. Website about airline food, via Yan Li

2. The economics of amniocentesis

3. Arnold Kling has an important update

4. Women and hot cars, etc. -- what science has to say

Why so many churches in Las Vegas?

Robert, a loyal MR reader asks,

I lived there for four years and was always curious.  I guess the more sin, the more churches???

For background, here is a 1997 look at the numbers.  It seems the city has more churches than average in per capita terms.  Among the cities claiming the highest number of churches per capita are Nashville, Grand Rapids, Mich., Waco, Texas, Wheaton, Ill., and Berkeley, Calif. (Berkeley?)  Overall the South has more churches per capita than the rest of the country.

I would think that most of the churches in Las Vegas are for the residents, not the tourists, and thus the quantity of sin is not a major factor.  Why should the residents be especially sinful?  (Don't forget that lots of sinful activities seem to be produced at constant returns to scale, so there's not always free-riding upon the sin infrastructure for tourists plus parking is an issue.)  In explaining the number of churches, I would expect three factors to play a role:

1. Perhaps migrants to Las Vegas are more likely to come from the South.

2. Most of the population growth is recent, so churches serve a valuable function of social networking.

3. Las Vegas has no dominant established religion so there is much religious competition and thus many different churches.

But surely Jacqueline (and others) can set us straight...

Silly me

I'd like to repeat how surprised I am at the size of the gains from drilling, or at least what appears to be the best current estimate of such gains.  What happened is simple.  I read people writing "Price won't go down much from drilling, therefore the gains are small," in various ways.  That is a non sequitur because there is producer surplus as well!  I committed a simple error.  Yet again.  And I committed this error because it fit in with my (nonetheless largely correct) preconception of politicians as bozos who will say anything for political gain or to sound good on TV.

Addendum: An Op-Ed on the topic.  And here's a skeptical Andrew Sullivan reader.  Like me, it doesn't seem he ever tried to look for a number.