A Real Stimulus Plan

As you know, I’m not enthusiastic about a fiscal stimulus plan.  What we need is a stimulus plan that does not increasing the budget deficit or waste taxpayer funds but that does increase the incentive to produce output.  So what would I do?  Here’s a new idea.

The IRS knows how much income that each taxpayer reported last year.  So let’s cut everyone’s marginal tax rate based on last year’s income.  In other words, suppose that last year Joe earned $66,520 which puts him in a 25% tax bracket.  Joe’s tax schedule this year will be exactly the same as last year except for every dollar earned above $66,520 the tax rate drops to 15%.   We do this for all taxpayers so that each taxpayer has their own schedule and for each taxpayer there is a decreasing marginal tax rate.

Note that this plan increases the incentive to work and it doesn’t increase the deficit.  In fact, the Tabarrok plan increases tax revenues!  The key is a marginal tax cut with a different margin for every taxpayer based upon last year’s return.

That’s the basic idea but there are some obvious modifications that could solve various problems.  For example, the new schedule could be based on an average of say the last three years of income or the average plus some roundup for growth etc.  It’s also possible to cap the base on which the lower marginal tax rate applies, for example, we could create a lower tax rate on every dollar of income above last year’s income up to an increase of 20%.

It is true that a permanent system like this could be (partially) gamed but the system can work very well if used occasionally, say for the most serious recessions.  We would also learn a lot by applying this system and looking at the taxpayer response. 


I am absolutely the dumbest guy that reads MR - and choose to comment rarely for fear of ridicule, but:

With the new marginal tax, does Joe just talk to his boss and say "I know there's a recession, and I know I'm salaried, but I'd like to work more and get a raise, please." Is the incentive of a lower marginal tax rate enough to make Joe go get a second job if that fails?

Seems vaguely Laffer-esque to me...

I like the idea but I am not sure that it won't decrease tax revenue? In a hypothetical world where the
total wage volume is given but its individual distribution fluctuates from one year to the other (ok,
that's not very realistic), the increase in taxes from the group of people who see their revenue
increase would be smaller than the decrease in taxes from the group of people who see their revenue fall.

Yeah, but under the Tabarrok plan all those people who are paying no taxes will not get a check in the mail! And of course, you are just giving another hand out to the rich. And maybe YOU, in your ivory tower office, have the option of expanding work effort such as delivering more guest lectures and writing more books, but what about the "typical" worker? Incentives don't really matter, do they?

Jeez, a half-second's thought gives you the way to game it.

Move the following year's payments into the previous year.

I think Bell Labs had a permanent oddity in payroll where there were two payments in December and none in January, or something like that, to take advantage of some minor one-time tax change or other in the 60s.

Aren’t taxes supposed to be progressive for good reason? You are turning the picture around. Give the one that have more. They know best what to do with it.

Good to see new ideas. However, I wonder if it's really the problem that people aren't working hard enough. Americans already work very long hours compared to the rest of the developed world. Is it sustainable for them to further boost their working hours and if not, won't there be a problem when they later cut back?

Brock's point also seems valid.

How would that apply to those currently in the 10% and 15% brackets? Just curious... doesn't seem like all that great of a deal to them.

Brock makes a good point, the problem with a recession is that there is less work than people are willing to do, leading to a decrease in spending which can potentially spiral the problem further down. So there is already underemployment, and all this plan is going to do is increase the amount of people looking for jobs that are not there for the moment.

However, there is a version of it for recessions: why not increase the tax rate for income higher than this years, and use the money for some form of stimulus? After all, people whose income increases during a recession are people apparently not hit by the recession, and they are the natural source to pay for a stimulus package.

I presume this is a common point with regressive taxes, but doesn't this plan increase the marginal cost of extra workers compared to making existing ones work harder? Making it a disincentive to creating extra employment.
And don't you think people in the US are working hard enough already and the social costs of encouraging them to work longer hours are worth considering? The US does already have longer working hours than most developed countries already doesn't it?

Not a bad idea, but two problems....

1) the IRS cannot keep up with IT programming now
2) I don't see any miraculous incentives in this

The highest earners have any number of deferral and avoidance possibilities already.


Many fail to recognize that tax deferral and avoidance are only second best strategies when a major chunk of earnings would be lost to taxes. When tax rates are reduced, the value of deferral and avoidance decline. I just finished Reynolds' Income & Wealth and I believe he cited studies showing an increase in taxable income of 40% when rates were cut just due to income leaving shelters. The taxes generated by the increase in declared income is the excess over what would be expected just assuming that taxes go down in proprotion to the reduction in rates. Also, that is before any increase in effort in response to higher after-tax income. Supply-side theories predict higher than expected revenues from two sources: reduction in the value of shelters and incentives to produce more. Disclaimer - I do not claim that tax cuts are cost free, at least in the short term.

I don't know how I feel about Alex's proposal, but I always love to see 'em complain about any suggestion that working people keep more of their income.

The tax doesn't cost any revenue because the drop in revenue would occur with or without the tax cut. The cut is designed to spur extra work, which could result in higher than expected growth and thus higher than expected tax revenues.

Under the scheme, companies would accelerate expansion plans, especially if the wage incentive on workers was coupled with a depreciation incentive for companies. Workers would also have an incentive to work more, perhaps proposing to accelerate their own schedules, proposing new expansion plans, or starting new businesses. In fact, some businesses may be forced to increase production, because their most able and talented employees would demand extra work and pay or they would shirk at their current job to focus on other work.

Many people are making the mistake of thinking that demand and supply side factors are completely distinct. A supply side incentive plan that encourages work also encourages more spending.

Also an incentive plan such as this must be compared with some alternative plan such as a fiscal stimulus.

I still don't get it. How does it "increase tax revenues" with an exclamation point? Are you invoking the Laffer curve effect or some other reasoning?

The professionals that would benefit the most from this tax plan would not be affected by the incentive here. For example, associates at corporate law firms receive a lockstep advance in salary of around 10-15% of their salary every year, in addition to any bonuses, regardless of how much more they work (than is already expected). Other professionals working on a partnership basis (lawyers, doctors) could increase the price of their services to take advantage of the tax plan. The end result would be tax "gifts" without any behavioral change and higher prices for consumers of some services.

If I understand correctly I thought that income effect and substitution effect roughly balanced, so that a cut in marginal rates might not incentivize more output accross the board.

So, I don't see how this provides a stimulus nor additional revenue but I'm just some guy in pajamas and could easily be mistaken.

Many supply side advocates forget that when Reagan took office marginal rates were, if memory serves me correctly, 70%.

They assume the Laffer Curve or whatever supplu side theory is going to work the same with today's marginal rates. No chance.

(I've worked with a lot of high end individuals, and none of them ever worked harder due to tax cuts. They did rethink avoidance and deferral strategies however.)

Also, rate discussions are misleading unless one understands the maze of deductions, exemptions and credits that change effective tax rates.

And the IRS still can't get its' computers up to date.

Alex, you've stumbled onto a project we've been developing for this April over at Political Calculations: a tool that you can use to design your own simpler tax code. One of the key features available to prospective tax code designers would be an income cap - a threshold above which additional income would either not be taxed or that would be taxed at a lower marginal rate.

As you've noted, if the government really wanted to encourage taxpayers to generate more income (and in doing so, seriously generate economic activity), while at the same time guaranteeing a rock-solid flow of tax revenue for the government to support vital programs, this is the way to go. It solves the incentive problem of high tax rates discouraging additional income-generating activity and the associated tax-avoidance strategies used by those who are able to control how and when they realize the money they make.

You'll recognize that this isn't a new idea. There's a reason why Social Security taxes were designed to function in a similar fashion (mainly to provide rock-solid and predictable levels of funding for the program), but the economic growth generating aspect has never been been able to be realized as extremely high progressive income tax rates have constrained the growth potential.

There's more features that our tool will explore, including negative rates of income taxes for low-income earners (someone's got to keep Milton Friedman's ideas alive!), but we're certainly open to other simple, yet powerful ideas....

How does this not increase the deficit, are you saying that because it's for one year only it won't increase the structural deficit?

"doesn't cut taxes on existing economic activity"

that is only true if you have a way of excluding normal variations in economic activity, and separating them from new incentivized activities.

Comments for this post are closed