Optimal Oil Production and the World Supply of Oil. By Nikolay Aleksandrov, Raphael Espinoza, and Lajos Gyurko
IMF Working Paper No. 12/294
Dec 2012
http://www.imf.org/external/pubs/cat/longres.aspx?sk=40169.0
Summary: We study the optimal oil extraction strategy and the value of an oil field using a multiple real option approach. The numerical method is flexible enough to solve a model with several state variables, to discuss the effect of risk aversion, and to take into account uncertainty in the size of reserves. Optimal extraction in the baseline model is found to be volatile. If the oil producer is risk averse, production is more stable, but spare capacity is much higher than what is typically observed. We show that decisions are very sensitive to expectations on the equilibrium oil price using a mean reverting model of the oil price where the equilibrium price is also a random variable. Oil production was cut during the 2008–2009 crisis, and we find that the cut in production was larger for OPEC, for countries facing a lower discount rate, as predicted by the model, and for countries whose governments’ finances are less dependent on oil revenues. However, the net present value of a country’s oil reserves would be increased significantly (by 100 percent, in the most extreme case) if production was cut completely when prices fall below the country's threshold price. If several producers were to adopt such strategies, world oil prices would be higher but more stable.
Excerpts:
In this paper we investigate the optimal oil extraction strategy of a small oil producer facing uncertain oil prices. We use a multiple real option approach. Extracting a barrel of oil is similar to exercising a call option, i.e. oil production can be modeled as the right to produce a barrel of oil with the payoff of the strategy depending on uncertain oil prices. Production is optimal if the payoff of extracting oil exceeds the value of leaving oil under the ground for later extraction (the continuation value). For an oil producer, the optimal extraction path corresponds to the optimal strategy of an investor holding a multiple real option with finite number of exercises (finite reserves of oil). At any single point in time, the oil producer is also limited in the number of options he can exercise, because of capacity constraints.
Our first contribution is to present the solution to the stochastic optimization problem as an exercise rule for a multiple real option and to solve the problem numerically using the Monte Carlo methods developed by Longstaff and Schwartz (2001), Rogers (2002), and extended by Aleksandrov and Hambly (2010), Bender (2011), and Gyurko, Hambly and Witte (2011). The Monte Carlo regression method is flexible and it remains accurate even for high-dimensionality problems, i.e. when there are several state variables, for instance when the oil price process is driven by two state variables, when extraction costs are stochastic, or when the size of reserves is a random variable.
We solve the real option problem for a small producer (with reserves of 12 billion barels) and for a large producer (with reserves of 100 bilion barrels) and compute the threshold below which it is optimal to defer production. In our baseline model, we find that the small producer should only produce when prices are high (higher than US$73 per barrel at 2000 constant prices), whereas for the large producer, full production is optimal as soon as prices exceed US$39. Optimal production is found to be volatile given the stochastic process of oil prices. As a result, we show that the net present value of oil reserves would be substantially higher if countries were willing to vary production when oil prices change. This result has important implications for oil production policy and for the design of macroeconomic policies that depend on inter-temporal and inter-generational equity considerations. It also implies that the world supply curve would be very elastic to prices if all countries were optimizing production as in the baseline model—and as a result, prices would tend to be higher but much less volatile.
We investigate why observed production is not as volatile as what is predicted by the baseline calibration of the model. One possible explanation is that producers are risk averse. Under this assumption, production is accelerated and is more stable, but a risk averse producer should also maintain large spare capacity, a result at odds with the evidence that oil producers almost always produce at full capacity. A second potential explanation is that producers are uncertain about the actual size of their oil reserves. Using panel data on recoverable reserves, we show however that, historically, this uncertainty has been diminishing with time and therefore this explanation is incomplete, since even mature oil exporters maintain low spare capacity. A third explanation may be that the oil price process, and in particular the equilibrium oil price, is unknown to the decision makers. Indeed, the optimal reaction to an increase in oil prices depends on whether the price increase is perceived to be temporary or to reflect a permanent shift in prices. If shocks are known to be primarily temporary, production should increase in the face of oil price increases. But if shocks are thought to be accompanied by movements in the equilibrium price, the continuation value jumps at the same time as the immediate payoff from extracting oil. In that case an increase in price may not result in an increase in production. Faced with uncertain views on the optimal strategy, the safe decision might well be to remain prudent with changes in production.
In practice, world oil production is partially cut in the face of negative demand shocks. The last section of the paper investigates whether the reduction in oil production during the 2008–2009 crisis can be explained by the determinants predicted by the model. We find that the cut in production was larger for OPEC, for countries facing a lower discount rate, as predicted by the model, and for countries with government finances less dependent on oil revenues.
Erskine Bowles, who is sort of a Democrat, met Wednesday with House Speaker John Boehner to help Republicans promote proposals to cut entitlements, as part of the “fiscal cliff” negotiations.
This is the right place for Bowles, who has long maintained a mutual-admiration society with House Budget Committee chairman Paul Ryan, R-Wisconsin. The former Clinton White House chief of staff has always been in the corporate conservative camp when it comes to debates about preserving Social Security, Medicare and Medicaid.
It’s good that he and Boehner have found one another. Let the
Republicans advocate for the cuts proposed by Bowles and his former
Wyoming Senator Alan Simpson, his Republican co-conductor on the train wreck that produced the so-called “Simpson-Bowles” deficit reduction plan.
After all, despite the media hype, Simpson-Bowles has always been a non-starter with the American people.
Last summer, at the Democratic and Republican national conventions,
so many nice things were said about the recommendations of the National
Commission on Fiscal Responsibility and Reform that had been chaired by
former Wyoming Senator Alan Simpson, a Republican, and Bowles that it
was hard to understand why they were implemented. Paul Ryan went so far
as to condemn President Obama for “doing nothing” to implement the
Simpson-Bowles plan—only to have it noted that Ryan rejected the
recommendations of the commission.
But, while a lot of politicians in both parties say a lot of nice
things about the austerity program proposed by Simpson-Bowles, there is a
reason why there was no rush before the election to embrace the
blueprint for cutting Social Security, Medicare and Medicaid while
imposing substantial new tax burdens on the middle class.
It’s a loser.
Before the November 6 election, Simpson and Bowles went out of their way to highlight the candidacies of politicians who supported their approach—New Hampshire Republican Congressman Charlie Bass, Rhode Island Republican US House candidate Brendan Doherty, Nebraska Democratic US Senate candidate Bob Kerrey. Bipartisan endorsements were made, statements were issued, headlines were grabbed and…
The Simpson-Bowles candidates all lost.
Americans are smart enough to recognize that Simpson-Bowles would stall growth. And they share the entirely rational view of economists like Paul Krugman.
“Simpson-Bowles is terrible,” argues Krugman, a Nobel Prize winner
for his economic scholarship. “It mucks around with taxes, but is
obsessed with lowering marginal rates despite a complete absence of
evidence that this is important. It offers nothing on Medicare that
isn’t already in the Affordable Care Act. And it raises the Social
Security retirement age because life expectancy has risen—completely
ignoring the fact that life expectancy has only gone up for the well-off
and well-educated, while stagnating or even declining among the people
who need the program most.”
On election night, Peter D. Hart Research Associates surveyed Americans with regard to key proposals from the commission. The reaction was uniformly negative.
By a 73-18 margin,
those polled said that protecting Medicare and Social Security from
benefit cuts is more important than bringing down the deficit.
By a 62-33 margin,
the voters who were surveyed said that making the wealthy start paying
their fair share of taxes is more important than reducing tax rates
across the board (62 percent to 33 percent).
But that’s just the beginning of an outline of opposition to the Simpson-Bowles approach.
To wit:
* 84 percent of those surveyed oppose reducing Social Security benefits;
* 68 percent oppose raising the Medicare eligibility age;
* 69 percent oppose reductions in Medicaid benefits;
* 64 percent support addressing the deficit by increasing taxes on
the rich—with more than half of those surveyed favoring the end of the
Bush tax cuts for those making more than $250,000.
Americans want a strong government that responds to human needs:
• 88 percent support allowing Medicare to negotiate with drug companies to lower costs;
• 70 percent favor continuing extended federal unemployment insurance;
• 64 percent support providing federal government funding to local governments;
• 72 percent say that corporations and wealthy individuals have too much influence on the political system.
AFL-CIO president Richard Trumka is right. On November 6, “The American people sent a clear message.”
With their votes, with their responses to exit polls, with every
signal they could send, the voters refused to buy the “fix” that Erskine
Bowles is selling.