Oil subsidies

This May 2011 piece (What's an Oil Subsidy?) by Nick Loris and Curtis Dubay took the issue of subsidies to the oil and gas industry head-on.  I've had discussions with Heritage over a number of years on the issue of energy subsidies, though primarily those dealing with nuclear and ethanol. I hope those discussions will continue, and broad support for eliminating expensive and damaging energy subsidies will require support across the political spectrum.

In the May piece, Loris and Dubay took some positive steps in explicitly targeting some oil and gas subsidies for removal, something I don't think Heritage would have been willing to do ten years ago.  But the authors' defense of many of the other subsidies listed was disappointing and incorrect.

1)  Section 199 Deduction to support domestic manufacturing was defended on the grounds that it goes to all manufacturing, not just energy.  Many subsidies benefit more than one industry, and the argument that it is therefore part of the "tax baseline" is always used to protect the subsidy and divert prying eyes elsewhere.  This claim should almost never be accepted outright.  Rather, more careful review to assess whether the "general" subsidy disproportionately benefits a particular industry, geographic region, or fuel is needed.  In some cases, the fine print on eligibility actually codifies far more favorable rules for some market participants than for others under what is supposedly a "baseline" subsidy.  In others, the rules may seem neutral but be of far more practical value to one industry than another.

On Section 199, I think the whole thing should be killed.  But if it remains and we are talking about oil and gas industry investments into extraction equipment or refineries, I tend to agree with Loris and Dubay that the spending is fairly similar in nature to investments in other sectors and shouldn't be singled out.  That is only part of the story, however.  As I noted in an earlier blog post, the argument breaks down when counting the endowment value of extracted raw materials as though it were manufacturing.  It is not, and should not be includible in the calculation of section 199 credits.

2)  Foreign tax credit (FTC).  Loris and Dubay inaccurately portray the rationale for modifying the rules on the FTC.  The reforms would establish a better system and help eliminate abuses that are particularly easy for royalty-paying extractive industries.  The authors' have characterized the discussion of FTC-related subsidies as aiming to kill the FTC entirely for oil and gas firms, a move they say (correctly) would result in the double taxation of foreign operations.  But full elimination of the FTC is not what is being discussed now.  Nor has it been the recommendation of any subsidy study I've ever done or looked at. 

Rather, the goal is to end the abuses:  where corporate tax rates on oil, gas, or mining are more than those levied on other corporations, it is quite likely that royalties (which are deductible against US taxes) are being disguised as corporate taxes (which, under the FTC, would receive the much more generous tax credit).  This should end; and if doing so drives up the effective tax rate on foreign producers of oil, gas, or other minerals because they have been gaming the system for decades, so be it.

3)  Expensing.  The authors argue for immediate expensing of all capital equipment.  This same argument is included in a more recent posting by Loris here.  I will defer a detailed discussion on whether immediate expensing makes sense or not for our economy, as I've asked Nick for more info on the empirical justifications for such a move so we can have a more useful discussion on the topic at a future time.  However, the argument put forth in the posts is that immediate expensing reduces the cost of capital, and that this outcome somehow justifies the policy of full expensing. 

Such an outcome is self-evident, and not at all relevant to the underlying policy issue.  Expensing is a subsidy to capital (you are increasing the speed at which multi-year capital may be written off from current year taxes, creating a time-value of money benefit similar to an interest-free loan from the government), and subsidies generally reduce costs or prices. Government loan guarantees, insurance caps, and investment tax credits all reduce the cost of capital as well.  But minimizing the cost of capital is not the primary objective of economic policy.  If fact, subsidizing multi-year capital can distort economic choices across sectors and between capital-intensive and capital-saving methods to provide similar goods and services.  This is not a good thing.

On the positive side, the way Heritage covered the issue implicitly recognizes that under our current system of capital depreciation, special rules for one asset class versus another do create differential levels of subsidy support.  Thus, if we were to decide not to shift to the full expensing system of taxation Heritage is advocating, Heritage should be supporting policies that properly match depreciation periods with the actual asset service lives as their fallback position.  This would eliminate the political use of depreciation periods to reward favored industries, enhancing tax system neutrality across sectors.  Unfortunately, their actual fallback position appears far more partisan:

All companies, including oil and gas companies, should be able to expense their full capital costs immediately. Until that critical change in the tax code is made for all businesses, Congress should retain all provisions that move the tax code in the direction of expensing  [emphasis added].

Since the special rules have generally gone to the most effective lobbies rather than to the most effective technologies, maintaining the status quo does little more than protect the incumbent industries.

4)  Percentage depletion.  The assessment of percentage depletion allowances mimics the issue framing taken by the American Petroleum Institute (it is simply an "alternative" way to write off assets).  This is simply wrong.  Heritage notes that:

A depletion allowance is analogous to depreciation and is appropriate when the quantity of the potential resource is unknown, such as the amount of recoverable oil from a well. Independent oil and gas producers use a depletion allowance to recover capital investments over time. This is also available to producers involved in mining, timber, geothermal steam, and other natural deposits. The depletion allowance for independent oil and gas producers is 15 percent of the producer’s gross income from its average daily production, up to 1,000 barrels of oil. While there is nothing wrong with percentage depletion in theory, the question is whether at 15 percent it is overly generous or, possibly, not generous enough and should be raised.

But there is something wrong with it in theory:  percentage depletion rules allows certain industries to deduct from taxes more than they have actually invested into an asset.  As a result, the policy is a clear subsidy.  End of story. 

The issue of allowable percentage depletion rates matters only in terms of how big the subsidy is for one mineral versus another.  Recall that what is measured by Treasury and JCT is not the percentage depletion allowance per se, but the excess of percentage depletion over cost depletion.  Firms can shift to cost depletion in any year it is more favorable, so if they are using percentage depletion (and they are, because it shows up every year in the federal tax expenditures budget) it is because they are saving money.

The size of the deposits or the market value of that production is irrelevant to basic concepts of depreciating investments over time. The authors are perhaps mixing up tax accounting and financial reporting.  For tax accounting purposes, only the cost of the investment matters.  Absent the gain or loss on the sale of a well or mine, the IRS shouldn't care what your asset is worth because in a normal industry it does not affect taxable income (though obviously with percentage depletion it does). 

Assets across many industries shift in value year-to-year.  And, as with oil properties, the estimates of the remaining lives and gross future production capacity are also volatile for nearly all businesses -- dependent on broader market conditions regarding demand, market prices, and evolving production technologies.  Factories can be shuttered in bad times, and re-opened (at a cost) if market prices change.  Changes in the operating environment can have dramatic effects on asset values.  A biotech company that had invested $100m into its production facility, for example, might see that asset surge in value if FDA extended the medical conditions for which its approved drug could be prescribed.  Conversely, the rise of a competitor would not change the plant, but would alter expected future cash flows, and therefore depress the market value of the facility.  Yet, absent a bankruptcy, neither case would change tax depreciation schedules one whit.  Substitute "oil well" for biotech facility, and percentage depletion deductions would surge in the first and drop in the second.  It makes no sense.

If the firm wants to track well depletion for financial reporting, no problem.  The data may help its managers and investors assess future market trends and investment needs.  But the appropriate method for tracking assets for internal financial reporting is no justification at all to keep percentage depletion subsidies in our tax code.

With pressure building in Congress to strip out at least the most obvious subsidies to oil and gas, Taxyapers for Common Sense has released a new tally of some of the major ones.  The report is useful in providing updated cost estimates, and for going beyond the narrow set of provisions that the legislation has targeted.  For example, they pick up the billions in losses due to negligence by the Minerals Management Service in structuring lease contracts for 1998 and 1999 that resulted in taxpayers getting no royalties at all for massive quantities of oil and gas in the Gulf of Mexico. 

There are some areas where I disagree with how provisions are characterized, however.

VEETC.  The single largest subsidy tagged in the report is the volumetric ethanol excise tax credit.  The arguments I've seen presented to include this as a subsidy to oil rather than to ethanol have been that (a) the subsidy is paid at the point of blending, and the blenders are often oil companies; and (b) the subsidy is totally unnecessary since blenders would have been required to use the ethanol anyway under the Renewable Fuel Standards (RFS).  The RFS mandate use of pre-set levels of ethanol in the nation's motor fuel supply. 

Both arguments are inaccurate.

• Regardless of what point in the value chain the credit is earned, it is earned only for using ethanol and skews markets towards this input rather than other blending agents or fuel extenders.  The economic incidence of any subsidy (who ends up with the improved economic returns) often differs from the point of payment, and often shifts over time as the relative market power of different parts of the value chain shift.  But the policy clearly provides government payments for using ethanol, not oil.  
• It is true that the tax credit and the RFS are, indeed, duplicative.  But this duplication plays out through the price system.  The larger the tax credit, the lower the incremental cost (as measured by the trading price of a compliance unit under the mandate called a "Renewable Identification Number" or "RIN") will trade for.  Proper accounting for ethanol subsidies would be the sum of subsidies under the RFS and the credit (plus other policies as well).  All of these support the use of ethanol.  But the existence of dual systems merely means the full subsidy cost needed to reach the mandated level of consumption is split between taxpayer costs (through the credit) and consumer costs (through RIN prices pushed through into fuel prices).  It does not mean that the entire credit is a windfall to oil companies.

LIFO.  Last-in first-out accounting is one of a range of inventory accounting methods allowed to all industries under US law.  During times of rising prices (including those due to inflation), LIFO approaches result in higher near-term tax deductions.  During times of falling prices, first-in last-out (FIFO) results in higher near-term deductions.  With flat prices it doesn't matter much.  Over a longer period of time, the tax impacts of large past price surges begin to subside.

Thus, during a run-up in oil prices LIFO will provide large benefits, but if prices stay high for awhile the benefits to the Treasury from banning LIFO will diminish because more and more of the investory will have been procured at the new, higher prices.  Estimates of the tax savings from eliminating LIFO in oil and gas were done during a time of rising prices, and reflected the time window during which the changes in tax revenues were highest.  However, this is a short-term surge, not one that will result in continued higher revenues, year-in and year-out.  In constrast, overdue reforms such as eliminating the silly percentage depletion rules would generate recurring subsidy reductions.

LIFO may well fall in order for the US to comply with international accounting standards (which bar LIFO), but I don't consider it a subsidy to any one sector today.

The other two "general" tax subsidies TCS listed were the Manufacturing Tax Deduction for Oil and Gas Companies and Deductions for Foreign Tax Credits that are really related to resource payments.  I believe there are strong arguments for at least substantial portions of both of these programs to be counted as subsidies to oil and gas, and would actually favor including that portion within the primary list of subsidies to the industry (TCS has included these in a separate table).  More on the Manufacturing Tax Deduction here.

The TCS report did not cover all subsidies to oil and gas.  Among those left are large subsidies to bulk transport of oil via our inland waterway system, subsidized financing and operation of the Strategic Petroleum Reserve, oil defense, a variety of other accelerated depreciation provision for oil and gas infrastructure, and massive shortfalls in fuel tax collections to finance the nation's interstate highway system.  Thus, I expect that even with the adjustments noted above, the aggregate subsidy levels to oil and gas would be higher, not lower, than what they have reported.

A recent post by Ken Cohen, head PR guy at ExxonMobil, laments the temporary shut-down of offshore oil and gas drilling and its impact on all of us -- not them, but us, through lower federal taxes, higher deficits, less jobs, and less security.

Cohen titles his entry "A trillion-dollar missed opportunity -- enough to pay the U.S. deficit," and then goes on to provide data contradicting his claim.  He references research done by ICF International for the American Petroleum Institute to that claims the foregone earnings to government by not leasing up our coastal oil and gas resources are $1.3 trillion over the lifetime of the projects.  But the $1.3 trillion deficit Cohen is pointing in his blog title as being offset is an annual figure.  Thus, if the foregone projects last 10 years, the income would be only 1/10th of the deficit over the relevant period; if they last 15 years, 1/15th, and so on.  Net out subsidies, and the contribution to federal stability shrinks further.

The ICF study is actually kind of a mixed bag for supporting Cohen's points.  Page 31 highlights growing oil security risks around the world.  The fact that the US offers such a stable regime for oil and gas operators like his firm simply underscores how much the feds are undercharging energy companies for the leases right now, especially by improperly boosting federal takes automatically as global energy prices rise.  The feds should perhaps charge even more for the access in the future.  Note that this conclusion on US undercharges is broadly in line with a Government Accountability Office analysis of the issue a couple of years ago.  GAO noted that "the U.S. federal government receives one of the lowest fiscal takes in the world" (page 2, emphasis added).

ICF's data on page 17 regarding industry investment into carbon mitigation strategies is also problematic for Mr. Cohen.  Upon first skim, the data suggest that the oil and gas industry have been good guys, taking the problem seriously and investing more in mitigation than anybody else.  OK; so it's not hard to beat the coal industry.  But $58.4 billion in total investments for nine years (2000-08 inclusive) isn't much to write home about either for one of the core industries linked to the climate change problem.  That's $6.5 billion per year for the entire oil and gas sector.  In comparison, ExxonMobil alone (which, we learn on page 33, is not even very big by international standards), had revenues of $477 billion in 2008 and $311 in 2009.  Big difference.

Even these limited investments seem a bit diminished once one starts to dig into what the industry has counted as carbon mitigation spending.   The largest share seems to be directed towards making their own operations tighter, such as reducing gas flaring and improving the efficiency of refineries.  This is certaintly nice to see, but belongs in the same category as WalMart boosting in-store energy efficiency.  It's a stretch to treat these investments as an illustration of industrial leadership poised to to address sector-related problems head-on.  In fact, much of this spending is likely for high return investments that should have been done years ago, and would have been had there been any price on carbon.

For years Armin Wagner and his team have toiled to pull together a consistent set of motor fuel prices from around the world.  Though the work may not get the attention showered each year on the the Economist's "Big Mac" index, the GTZ analysis in my mind is far more important. 

GTZ tracks what is essentially a standard product around the world: gasoline and diesel vehicle fuel.  Yet, the range in prices for fuel consumers varies enormously across countries, spanning three orders of magnitude.  The chart below, covering 2008 data on gasoline, illustrates this striking span.  A link to a higher resolution version helps with readability and provides a similar display for diesel.

Retail prices for gasoline, by country, 2008

Yes, there are a few (somewhat) logical explanations for portions of this variation.  Remote countries pay more for transport and see higher prices.  Large oil producing nations such as Venezuela and Iran use domestic subsidies to fuel to buy off the domestic population in a fossil-fuel variant of "bread and circuses."   Some European nations use high petrol taxes to fund a variety of non-oil related activities.  Other nations see very high prices due to corrupt central governments combined with inadequate domestic oil supplies or refining capacity or access to imports.

But these explanations only go so far.  Note how close to the bottom of the spectrum the United States is.  Efficient markets?  Taxes only high enough to cover activities supporting petroleum-related issues such as leaking underground tanks and financing the roads on which these vehicles drive?  My general impression had been that at least in terms of the US highway system, users more or less paid their cost through fuel excise taxes. There were clearly cross-subsidies (normally to heavy trucks) and a variety of other subsidies directly to oil, but at least the cost of roads was being picked up.

Turns out that this is not even close to being true.  A detailed evaluation of federal transport subsidies recently released by the Pew Subsidyscope project (disclosure -- I'm on the project's advisory board) found that fees on road users, including fees on fuels, vehicle registration fees, and tolls covered only 51 percent of the nearly $200 billion the US spent on highway construction and maintenance in 2007 -- the lowest contribution from users since the interstate highway system began in 1957.  The rest of the funds came from non-user funding sources, such as general tax revenue and bonds (often subsidized tax-exempt bonds).  

Pew rightly points out that some motor fuel excise fees are directed to non-road uses, such as mass transit.  However, even if all of these fees are also credited to roadways, the user-fee share rises only to 65 percent -- still dumping more than a third of costs onto others.  And it is useful to note that the Pew analysis doesn't pick up everything.  State, county, and local roads are major expenditure items for most of sub-national governments, and not generally funded by user fees.  Even National Forests make payments in lieu of property taxes to state governments; federal roads to my knowledge do not. 

Yet even ignoring these other subsidies, the Pew results are striking in an era where countries are meeting to address options to curb carbon emissions.  Road transport is not only a significant source of greenhouse gas emissions in the US, but it is the country's primary energy security challenge.  Continued subsidization of the fuels directly, or the instrastructure that drives that consumption, makes little sense.  

 If we charged motor fuel excise taxes high enough to cover the cost of building and maintaining our roads (levied not only on petrol fuels, but ethanol and other transport fuels as well), US pump prices would be much closer to those in some European countries.  And while raising fuel taxes is considered the "third rail" of US politics (touch the issue and you will die politically), the Pew analysis demonstrates that the issue is not really about raising fuel taxes at all.  It is about removing subsidies.