Search This Blog

Sunday, November 01, 2009

The Burden of Taxes

A common belief is that levying additional income taxes on those with high incomes has negligible adverse impacts on the poor. In this post, for a hypothetical economy, I will show why this isn't true. In future posts, I will discuss how this result for the hypothetical economy still has at least some relevance to the real world.

Assume the following:
1. All people have exactly equal talent and capabilities.
2. All markets are perfectly efficient and fair.
3. Return on Investment (ROI) for all occupations is exactly equal, where the definition of "Investment" is non-standard and includes the following:
a. Cost (Money & Opportunity Cost) of Education.
b. Difficulty/Stress/Unpleasantness of the job.
c. Negative of the Satisfaction produced by the job.
d. Other similar considerations.
Note that (b), (c) and (d) have a strong subjective component and even (a) has a significant subjective component via an individualized Discount Rate (a rate of interest that relates the value of future income to current income to a given individual). Because of the differing "Investments" required for various occupations, there would be significant differences in compensation for those occupations since the ROI for all occupations is constrained to be equal.

This means that only those occupations that provide adequate value to justify the ROI will exists. In other words, if an occupation requires a great deal of "Investment" and therefore requires a high wage according to the ROI constraint, but nobody is willing to pay for the goods or services that this occupation would provide at that wage level, the occupation won't exist.

The ROI is an after-tax ROI. In this hypothetical world, nobody cares what their compensation is before taxes and instead they focus on their after tax compensation.

Consider the effect of increasing taxes on a given occupation while leaving the taxes on all other occupations the same. It doesn't matter if it's an occupation with high or low compensation, the effect is the same. If everybody with the occupation stayed in it, the after tax ROI for that occupation would be reduced. However, this violates the ROI constraint, so enough people would leave this occupation and do something else until the supply of people working at this occupation is reduced enough so that supply and demand balance to drive the after tax ROI for this occupation back up to nearly its original level.

Since people have left this occupation and are now competing in other occupations, all occupations' ROIs are slightly reduced (including this occupation with its additional tax burden). Taxes and regulation essentially have the effect of reducing our ROI for the work we do.

That's the first interesting effect. All ROIs are reduced by the same amount. Any attempt at making income taxes progressive is completely thwarted. The ROIs for the occupations with the highest and lowest compensations remain equal after the tax increase.

Since the occupation with the new tax burden has similar after tax compensation, the pre-tax compensation must be higher. That means that the production of goods and services dependent on this occupation incur greater cost, which implies that everybody will end up paying a higher prices for these goods and services. Again, any attempt at making income taxes progressive is lost, since everybody, rich or poor, will pay the same increased cost for these goods and services.

In this hypothetical economy, it is impossible for income taxes, regardless how progressive by design, to actually be progressive in effect. Since the poor feel the burden of reduced ROI and higher prices most acutely, the poor feel the burden of higher taxes more than anyone else.

In future posts I'll argue why the real economy bears significant resemblance to this hypothetical economy in regards to taxation.

13 comments:

erp said...

Excellent analysis, but everything, even logic, has become politicized.

Low taxes bring in the most revenue, create jobs, etc., and high progressive taxes do the opposite, but the left doesn't care. They want to engage in class envy and punish the "rich."

Bret said...

Yes, but the poor aren't necessarily the "Left".

In fact, in talking to people I know who are less well off, I've found that they intuitively grasp that taxing the rich may make them worse off. They just don't know why. I'm trying to put together a framework to discuss how the poor might actually suffer more from high taxes then the rich.

If the Left loses the poor as a constituency, then the Left collapses, and the U.S. will become free once again.

Bret said...

Please note that this post is a work in progress and I intend to update it as I get feedback. I'll identify any changes I make here in the comments.

erp said...

The most obvious reason the poor are worse off is that the rich won't risk their capital creating new jobs or even maintaining the old ones if their profits are confiscated.

If the Left loses the poor as a constituency, then the Left collapses, and the U.S. will become free once again.

Won't that be a happy day.

Hey Skipper said...

I think there are some problems with your analysis.

First, decreasing marginal utility. The utility of, say, an additional $500 to someone in the lowest 5% of earners is far, far greater than to someone in the highest 5%.

Second, the components of investment may well be have negative correlation. Occupations that have a high cost of entry strongly tend to be less unpleasant and more satisfying. Therefore, the sum of a, b, and c might decrease even as a increases.

Now, that might hold if all people have identical talents and desires; however, since that is manifestly and irrevocably not true of the real world, that assumption has a whiff of warmenism (i.e. "Assume all feedbacks are positive." "But what if they aren't?" "Shut up. Assume all feedbacks are positive."). An assumption that is both pivotal and dubious is a tough sell.

Then there is the problem of whether difficulty/stress/unpleasantness and dissatisfaction even belong with investment in the first place. Given two jobs with an identical wage, the one that is more pleasant and satisfying is more valuable than the other. Consequently, I think your b & c belong in the return, not investment, bin.

Finally, depending on what you mean by "negligible", I'm not sure the math works. For the poor, the sum of services received and wages minus prices would have to be positive as the tax system becomes less progressive. Given the huge decrease in marginal utility as income increases, I'm not sure that would be the case.

Hey Skipper said...

One more thought. The far bigger problem with a highly progressive tax structure is that the government's income stream becomes very unpredictable.

As California is proving.

Howard said...

The utility of, say, an additional $500 to someone in the lowest 5% of earners is far, far greater than to someone in the highest 5%.

Even for a seemingly commonsense utility theory, the value of additional income to a high earner might be more valuable to other people if it translates into investment and job creation. I have participated in that role many times. However, utility theory only allows for the ordinal ranking of preferences for one individual at a time, not the comparison between individuals. Such misapplication is quite common.

Harry Eagar said...

Curiously, I occupied my morning drive asking exactly the same question, only about CIT.

It shouldn't exist, and now it doesn't. What's the harm?

Bret said...

Hey Skipper wrote: "The utility of, say, an additional $500 to someone in the lowest 5% of earners is far, far greater than to someone in the highest 5%."

Possibly (see Howard's comment).

In this case, the ROI is also higher for incremental compensation for lower paid occupations and this would tend to push compensation into a somewhat narrower range. I don't think it affects the analysis.

Hey Skipper wrote: "Second, the components of investment may well be have negative correlation."

Sure, but why does that matter? Think of "Return" being anything good about the occupation including compensation, job satisfaction, etc. and "Investment" being anything negative such as cost for education and training, time and lost opportunity costs, unpleasantness, etc. Each occupation for a given person still has a single, subjective "ROI" (though not specified in traditional units). Since each person can do anything, they're going to maximize that ROI based on market pricing and personal preference. An equilibrium (or random fluctuations about an equilibrium) will occur, where nobody can get a better personal ROI, because if they could, they would.

Hey Skipper wrote: "An assumption that is both pivotal and dubious is a tough sell."

The point of this post was to start somewhere. Things like this are very complicated and I often like to start with extreme simplification to start with a base agreement. If we can't agree here, then there's not much point in bothering to discuss the much more complicated real world.

Further posts will discuss what happens as these way overly restrictive assumptions are relaxed.

Hey Skipper wrote: "For the poor..."

I'm not able to figure out what you're trying to say in this paragraph and how it relates to anything in the original post. Could you try again?

Hey Skipper said...

Howard:

Even for a seemingly commonsense utility theory, the value of additional income to a high earner might be more valuable to other people if it translates into investment and job creation.

At some point, as the Scandinavian countries have proven, taxation becomes so progressive that it makes additional risk or effort a fool's errand.

It is also true that the additional taxes we already impose on high income earners have not reached that point.

What this shows, though, is that Laffer was correct: tax rates have an inflection point which maximizes the government's take.

The point in question here is whether, in our real world economy, making the US tax system less progressive would, through a more dynamic economy, yield more income to the poor than they lose to a greater tax bite out of their wallets.


However, utility theory only allows for the ordinal ranking of preferences for one individual at a time, not the comparison between individuals. Such misapplication is quite common.

Perhaps, but that is not how I am using it. Look at it from the opposite direction: for someone who already has a high income, the utility of even more income probably isn't the primary reason to engage in job creating activities.

Of course, you mentioned investment as part of job creation. IMHO that raises the more fundamental (and easily handled) question. Since the wealthy (or any of the rest of us, for that matter) can only do two things with money -- spend or save it -- is the economy better off if the wealthy have more of their money to spend and save, or if the government spends it instead?

Hey Skipper said...

Bret:

Since each person can do anything, they're going to maximize that ROI based on market pricing and personal preference.

Okay, I think I understand your point better.

I think it would be clearer (well, to me, anyway), though, by explaining ROI differently.

-- Investment is the combination of all the costs, money, effort & opportunity, to get into an occupation.

-- Return is total compensation, which is an occupation's combination of wage, difficulty, and satisfaction.

ROI would then be a dimensionless number, and a rational actor would select the occupation that maximizes that number (greatest Return for the least Investment).

Consequently, government policy that acts to diminish one of the components of Return will change individual decisions towards occupations requiring a lower investment.

Not only does that make intuitive sense, it also is reflected in the real world. Medical malpractice suits have had a particularly large impact on obstetricians, with the consequence that their Return, just as with an increase in taxation, has significantly diminished, thereby decreasing their ROI.

Consequence? Many fewer obstetricians in areas most friendly to tortious patients.

Which, in turn, means everyone, including the poor, pay more for obstetrics, and have less money to spend on things that benefit the poor more than paying lawyers and insurance companies just so some people can play judicial lottery.

Hey Skipper wrote: "For the poor…"

I'm not able to figure out what you're trying to say in this paragraph and how it relates to anything in the original post. Could you try again?


This relates to your thesis: A common belief is that levying additional income taxes on those with high incomes has negligible adverse impacts on the poor.

Presume that increased taxation on the wealthy translates into two things: higher EITC and fewer jobs (or, barring the minimum wage, the same number of jobs for less money) for the poor.

If an additional increment of taxation on the wealthy provides more income transfer to the poor than it costs in wage income, then the poor will, on net, gain.

This ties into the question of marginal utility: is the utility of the additional income lost to taxation sufficient to change the job creation behavior of the wealthy such that the increase in income transfer is less than the loss of jobs and increase in prices?

Bret said...

hey skipper wrote: "If an additional increment of taxation on the wealthy provides more income transfer to the poor than it costs in wage income, then the poor will, on net, gain."

Not in the hypothetical situation I've presented.

If an occupation gets a negative tax (for example, because of income transfer), than the ROI goes up, more people choose that occupation driving compensation down until ROI is identical for all again.

Again, I'll get to real-world differences from this hypothetical case in future posts.

I will shortly modify the post to take some of your comments into account. Hopefully, that'll make it clearer. Thanks for your input.

Harry Eagar said...

Another view