Sunday, May 3, 2009

Comparing the FairTax to the Income Tax -- Does the FairTax Depress Consumption?

The FairTax is a proposed national retail sales tax that would be designed to replace the income tax in the United States. It is explained quite well in the Wikipedia entry . FairTax proponents claim that a sales tax rate of 23% ($23 of every $100 spent) would be revenue neutral. Detractors claim that the 23% rate really amounts to a 30% sales tax (an increase of $23 in the after-tax cost for every $77 of goods and services).

A sales tax appropriately puts the tax burden on those who are using/consuming society's resources (whether natural or man-made), rather than on those who are creating value in society. You can probably guess that this blogger is a proponent of such a tax.

It's a good thing (or at least a fair thing) that such a tax hits those who consume the most, but we don't want to punish consumption too much. We want people to enjoy a high standard of living, and clearly too little consumption can cause near-term dislocations in the economy. Americans have historically not had a problem with being too frugal in the aggregate. The Japanese have, however, and perhaps a sudden jump to a very large sales tax could have negative consequences for the Japanese economy or a similar high savings rate, export-driven economy.

Does a sales tax really depress consumption more than the income tax, however? I think probably not. It is important to understand that under a primarily income tax-based system consumption is done with after-tax dollars.

Take the following example of a sole proprietor who has flexible work hours and finds himself in a situation where his marginal return for an extra hour of labor is $100. Under the income tax, he chooses to work an additional hour, makes an additional $100 and is taxed at a marginal rate of 25%. He will have $75 left after tax, and if he spends that money on dinner for four at a Chinese restaurant, he and his family will be well-fed for the night but will have no extra money left over from his hour of work. He has exchanged an hour of labor for some Chinese food.

Now let's consider the same situation under a sales tax similar to the proposed FairTax. We will assume a 25% tax rate using the FairTax definition of the tax rate (i.e. prices of goods and services are gross of taxes, and the amount of the tax is 25% of that gross price). Under the sales tax system, the proprietor will have $100 after tax. The Chinese dinner will cost him $100 ($75 price + $25 tax), and at the end of the day, he will be in the same situation as under the income tax.

FairTax proponents have come under criticism for claiming that the sales tax is only 25% in this example. Critics claim that the tax is really $25/$75 = 33.3%. I certainly agree with the critics that the sales tax is 33.3% and not 25%, but the FairTax people are also correct in pointing out that the income tax is never calculated this way. It is very easy to see in the example that a 33.3% rate of sales tax is equivalent to a 25% rate of income tax. It would be nice if we could get the government to admit that when somebody works to make $75 of after-tax income and pay $25 in tax that perhaps the real tax rate is 33.3% and not 25%.

One more example:

Let's suppose our businessman is also an amateur plumber, and he needs some plumbing work done in his house. He is more productive as a businessman than as a plumber, but he can hold his own against quality plumbers in the local area. Plumbers in the area make $75/hour, so it makes no sense for him actually to work as a plumber.

If he hires a plumber at $75/hour to do the required work in his house, he has to pay $100/hour under a 25% sales tax system. By doing the work himself he saves himself $100/hour, which is exactly what he would have been making doing his regular job. He is indifferent to doing the plumbing work himself at a 25% tax rate, but if the tax rate were higher, he would probably spend his time on the lower productivity plumbing job rather than on his higher productivity regular job.

Clearly, this is an inefficiency. It would be nice if the businessman did what he does best, and the professional plumber got to do more work doing what the plumber does best, but because there is no sales tax charged on work you do for yourself, there is an extra incentive to be a do-it-yourselfer.

If you do the same analysis under an income tax, however, the results are identical. In this case, there is no income tax charged on work you do for yourself, so there is the same incentive to be a do-it-yourselfer.

We see that the income tax discourages both production (because you get less money for your work) and consumption (because you're paying for your consumption with after-tax dollars). The sales tax/FairTax depresses consumption because it directly increases the cost of goods and services. But does it discourage work?

A blinkered economist might argue that it does. The utility of the dollars you earn is lower because their consumptive value is lower, so you would be getting less value for your labor. I think in practice, however, the decrease in the utility of dollars earned is mitigated by the fact that you earn more on your savings. If you can earn a high return on savings, it encourages you to go out and accumulate money with which to invest. People do tend to value having money in the bank, anyway, even if they have no plans to spend it. Overall, I think it is pretty clear that work is discouraged a lot less under a Fair Tax than under the current income tax.

An important side effect is that productivity will rise and the cost of goods and services will drop. So, even though the Fair Tax directly increases the cost of goods and services to the end consumer, the increase in productivity might very well go a long way towards offsetting that increase. There is no such offset under the income tax.


David said...

Ok, this is basically on-topic...
It seems that economists are confused about macro. It's a little weird that this has persisted for decades, but I suppose that macroeconomics is a relatively new field.

I was just listening to a podcast (here) of Russ Roberts interviewing Steve Fazzari about Keynesian economics.

Fazzari was explaining that recessions are caused by a reduction in "spending". He gave an example of a family that decides not to eat out on its usual night and to save the money they would've spent.

Roberts (a non-Keynesian) replied that although this would lead to a decrease in consumer spending, investment would increase by the same amount. So everything is hunky-dory.

Fazzari countered that while the family saved (say ) $100 in the process, the restaurant's savings dropped by the same amount, so net savings and thus net investment was unchanged. But consumption dropped. So it's a net loss.

Now am I imagining things? This still seems wrong.

Let's say that the family directly lent the $100 they saved to IBM so IBM could conduct some research. So economic activity simply got shifted from the restaurant to IBM, so it seems like there's no "harm". You could argue that if everyone did this at once, then there'd be a painful disruption in the economy, and I'll grant that. But that's not what the Keynesians argue.

On the other hand, if the family bought treasuries with the $100, then net consumption on that day would indeed be lower by $100. You could argue that that's "bad".

So if the government wants to "do something" about this, they can just go out and spend an extra $100 or cut taxes by $100. In other words, the gov't would have to have to run a higher deficit as the public bought more treasuries. Is this right?


ESM said...


You are pretty much correct. This guy Roberts is confusing investment with saving. Investment is a form of spending. It just happens to be spending that results in something of lasting value, but it's still spending. When the family doesn't spend, that's called saving. Sometimes saving can make it easier for somebody else to spend because the cost of borrowing drops a touch. In your example, IBM took advantage of the lower borrowing costs to invest (which is the best kind of spending of all). Replacing consumptive spending with investment spending is certainly a good thing in general, but of course we do want people to enjoy life too.

In your example, suppose borrowing costs were not lowered enough to incentivize IBM to invest $100 in research. In that case, economic activity has slowed because of the family's savings. Aggregate demand has dropped below the aggregate capacity of the economy to produce goods and services. In this case, it makes sense for the government to cut taxes until aggregate demand increases (as it surely will as the net financial wealth of the private sector increases).


David said...
This comment has been removed by the author.
David said...
This comment has been removed by the author.
David said...

In your example, suppose borrowing costs were not lowered enough to incentivize IBM to invest $100 in research. In that case, economic activity has slowed because of the family's savings.
But why wouldn't borrowing costs drop enough?

Thanks for the explanation.


ESM said...


It would be a miraculous coincidence if the extra $100 of savings lowered the cost of borrowing by an amount that generated a precisely offsetting amount of spending/investment.

First and foremost, there may be a non-zero gap/spread between the rate at which IBM can currently borrow and the rate at which they can profitably borrow. And $100 may be too small to close that gap. Remember, the private sector demand for money does not have to match the private sector supply of money because the government stands ready to absorb all excess money (or satisfy all shortages) at an interest rate set by the Fed. Second, even if you want to argue that at least the spread between the risk-free rate and the rate at which IBM can borrow should be reduced by the excess saving, I don't think this is true since a marginal increase in the aggregate desire to save does not necessarily mean a marginal increase in the aggregate desire to accept risk. Finally, there are still lags and frictions in the system which prevent the excess saving from translating directly into lowering borrowing costs. In the family saving example, maybe the guy who was planning to pay with $100 of cash just leaves the money in his pocket for a few days and even forgets about it.

A key point is that the response of the aggregate desire to spend versus the aggregate desire to save is pretty hard to predict when you are tinkering with interest rates and/or net financial wealth. You can be pretty sure of the sign of the effect, but the magnitude of the effect is probably beyond the predictive capacity of current financial models and is probably highly non-linear to boot.

Anonymous said...


I'm late to the game, but having just read this blog entry, I note that you didn't address my biggest concerns WRT the FairTax. One is that there will be a large hiccup in the economy as the public comes to grips with having to pay significantly more for goods before realizing or understanding the offsetting income adjustments.

The other is that unless the 16th Amendment is repealed at the same time that the FairTax is instituted, it is highly likely that the Government will only temporarily suspend collecting income taxes. Thus, we'll get a national sales tax and income taxes!