Monday, June 27, 2011

A Possible Budget Compromise

My first preference, of course, would be to balance the budget entirely by cutting expenditures; Reason has had past discussions on how that could be done. The basic problem, in my view, is not the deficit, troubling although it might be, but the amount of resources consumed and misallocated by government.

But if there is going to be a compromise involving increasing taxes and decreasing expenditure, I think the right way to do it would be to make the "tax increases" take the form of reductions in tax expenditures, the elimination of features of the tax code whose only purpose is to subsidize one or another activity, and which distort economic decision making in the same way as other subsidies do. The obvious big one is the deduction for home mortgage interest, which subsidize home ownership relative to renting and thus played some role in creating the recent financial crisis—although I suspect that is politically untouchable.

The Obama administration, in theory, seems to agree with that approach; a recent news story claimed they were proposing to eliminate a large tax expenditure associated with an accounting rule. Unfortunately, the news story I saw on the proposal got the substance of the rule (permitting a choice between LIFO and FIFO accounting) completely wrong, making the difference look both larger and less justified than it is. As I explained in my comment to the story.
"He used the example of an oil company that bought oil when prices lower and sells it when the price is higher, declaring its profit based on the higher price. "We just don't think that's right," he said."

It isn't right--and it has nothing to do with the LIFO/FIFO issue. Either the spokesman is lying, he is incompetent, or your reporter got confused.

The difference has to do not with the price when sold but the prices at which two different batches of the same product, such as oil, are bought. A firm buys a thousand gallons of oil at $2/gallon, puts it in a tank. Six months later it buys another thousand at $3/gallon, adds it to the same tank. Six months after that it sells a thousand gallons out of the tank for $4/gallon.

Under LIFO ("Last In First Out") the oil sold is considered to be from the second batch bought, so cost is $3/gallon, revenue $4/gallon, profit $1/gallon. Under FIFO ("First In First Out") it is considered to be from the first batch, profit $4-$2=$2 gallon. Note that the ultimate effect is on the timing of the profit, not the amount, since the second thousand gallons will eventually get sold too, and the total expense for the whole two thousand is the same either way.

Under neither rule is the cost set to the sale price, which would make profit zero and is what your story seems to be claiming.

For what it's worth, I teach this as one bit of a course at the law school of Santa Clara University.

7 comments:

$9,000,000,000 Write Off said...

A flatter, more neutral tax system is better, but I don't use the "tax expenditure" terminology. That assumes that the government owns 100% of your income and every dollar it lets you keep is some kind of indulgence on its part.

I prefer calling such deductions and credits "indulgences". Indulgences fund the operation of both sects; Politician X telegraphs that he is targeting a credit, and then Johann Tetzel visits its beneficiaries to raise funds for Politicians Y and Z, who will continue the indulgence. Next time Politician Y threatens a deduction and Tetzel collects for Politicians X and Z, and so on.

Exhibit A is Warren Buffet's tireless efforts (and bottomless funds) to expand the estate tax and thereby sustain his life insurance companies' primary source of revenue.

David Friedman said...

I expected some libertarian criticism along the lines $9... offers. I'm not basing my arguments on moral criteria but on economic ones. If expenditures on rent are not deductible and expenditures on mortgage interest payments are, that gives people in search of housing an incentive to buy even when, taking all costs and benefits into account, it would be better for them to rent. That is an undesirable consequence, at the same time that the government getting less money is a desirable consequence.

Thomas L. Knapp said...

I'm generally a "now now now" type of person, but it seems to me that just outright repealing the home mortgage interest deduction would further increase the default rate in the short term and make a bad situation worse in a bigger way.

Better to sunset it, with existing mortgages taken out under the deduction regime enjoying that deduction, but not new mortgages or re-finances. That would still probably further depress new home buying and construction in the short term, of course, but hopefully the produced correction would be less catastrophic and for fewer people.

$9,000,000,000 Write Off said...

I suppose the objection to terminology might be libertarian, but the point is a practical one.

Everybody at JCT, Ways and Means, the ABA Tax Section, and local analogs already know that eliminating special credits and deductions flattens the tax and produces a better tax system. No one disputes the merits of what you describe, but all legislation preserves or creates more distortions. Why?

Maybe the Martin Luther analogy is a stretch, but the answer is there.

Anonymous said...

I'm all in favor of eliminating some special tax breaks and subsidies intended to encourage some particular kind of economic activity. There was a brief glimmer of hope that the oil-company subsidies (which cause oil to be consumed at above the efficient level -- obviously in the public interest!) would get the axe, but between oil-state legislators, automobile-state legislators, and lots of drivers who vote, it was politically dead.

Each of these subsidies and tax breaks benefits somebody, and that somebody has more incentive to defend the subsidy than any other individual stand to gain by eliminating it -- so the former usually wins the lobbying battle. (The oil subsidy is actually unusual in this regard: it benefits a lot of people in the short run, at the expense of everyone in the long run. I think a more common situation is a subsidy that benefits a few people a lot, at the slight expense of all taxpayers.)

So how could you possibly beat this political calculus and eliminate a subsidy? I think the answer is not a "sunset" but rather a "dusk": not cutting off the subsidy at a particular date in the future (which predictably produces an enormous lobbying effort just before that date to extend it), but rather a gradual reduction in the size of the subsidy, so it reaches zero after several years. This not only produces less of an economic shock when the subsidy does end, but also dilutes the lobbying power of those who benefit from the subsidy by spreading it over time. Even if they mount a successful lobbying campaign this year to extend the subsidy, a they get is a one-year delay in the loss of, say, 20% of the subsidy, and then they have to do it all over the next year. The cost of lobbying is unchanged, but the reward has been reduced sharply.

Anonymous said...

I'm curious why you find the deficit troubling to any degree. Given that the U.S. isn't on a gold standard, it can emit liabilities without worrying that investors may make claims in the future.

As for tax deductions...surely the most noxious one by far is the state tax deduction?

Dick White said...

Let's assume the budget compromise isn't evolving because the politicians just couldn't marshall the willpower to support a compromise.
We're told of the Armaggedon effect of no debt ceiling increase on markets, credit and otherwise. It is this spectre that presumably forces the politicians to approve a ceiling increase without a corresponding structural change in the discretionary and non-discretionary budgets.
What if there is no ceiling increase and Armaggedon occurs. Might not the resulting pain provide politicians the cover, as it were, to make the necessary structural changes.
Current citizens would undoubtedly be harmed but perhaps future citizens might be relieved of some liability that otherwise will fall to them.
Is there a reason other than the ostrich effect, why this course of action is not considered?