Garett Jones  

Tax Rates, Efficient Government, and Jobs: Prescott's Surprise

PRINT
Romney Tax Cuts: Income and Su... Obama's War on Coal?...
"Why do Americans work so much more than Europeans?" That's the title of one of Nobelist Ed Prescott's papers.  His story has something to do with high taxes causing low employment, so you might be tempted to begin your yawning momentarily: Old news.  

But don't do that.  Prescott rejects one of the main tenets of free-market thinking--that the welfare state is a big waste.  That rejection is the key to his Big Idea.  

David's post on Governor Romney's "lower-rates/broader base" tax reform is what reminded me of this idea, and you'll realize there's a lot of overlap.  But Prescott takes the tax debate in quite an unusual direction.

Here's the story: Whenever government raises the tax rate on a particular person, an economist's first instinct is to think that the tax hike has two effects:

1.  The higher tax rate makes it less appealing to work in the formal market economy (this is called the substitution effect). With higher taxes, early retirement, unpaid work at home, and under-the-table work all look like better options.  

2.  The higher tax rate makes you poorer (the income effect). And poorer people (other things equal) would prefer to work more.  If your retirement gets wiped out in a market crash, you might delay retirement a few years; same thing if your tax rate goes way up.  

So when you ask a typical labor economist what a tax hike does, they'll have to think to themselves: Does the "I feel poorer" effect balance out the "Work pays less" effect?  

Answer: It depends.  

But Prescott's not a labor economist: He's a macroeconomist. That means he (implicitly!) asks a big question, oft-neglected by the labor folks: "What does the government do with the extra tax money?"  

You know the conventional response of free-market economists: "Government wastes it!  It gets burnt on endless boondoggles!  The government just grabs potatoes and throws them in the river!"

If that's a fairly accurate summary of government spending, then it's hard to tell whether a tax hike raises or lowers labor supply.  The government distorts work incentives (so you want to work less) and then destroys the output it grabs from you (so you want to work more).  Total effect on work effort?  We'll need to bring in the econometricians to figure out if it's a "+" or a "-".  

But Prescott rejects the free-market response: Instead, he takes the claims of welfare-state supporters at face value.  He takes it for granted that government output--health care, retirement funds, job training, all the rest--is just as awesome as privately produced output.  A foolish, foolish assumption, no?  A grave concession to the opposition, no?   

No and no. 

Here's the Wisdom of Prescott: If the government raises the tax rate on you and all of your friends, and then divides up all the tax revenue and dumps it from a helicopter, what do you get?  Well, none of the money gets wasted, so the tax hike doesn't have a direct income effect (there's a small indirect one I'll ignore here). 

If the tax hike is used for pure redistribution from the "average person" back to the "average person," then the tax hike doesn't make the "average person" poorer: The government is taking money out of everyone's right pocket and slipping it into their left.  

But if the income effect is gone, what's left?  The disincentive to work: The pure substitution effect. 

So here's Prescott's Big Idea: 

If higher taxes are wasted, then a tax hike has a small, ambiguous effect on employment.

If higher taxes are spent wisely, then a tax hike causes a big fall in employment. 

Not quite what you expected, was it?  

Coda: I'll save the data for another time. 


Comments and Sharing





COMMENTS (44 to date)
Richard Manns writes:

No doubt my ignorance has introduced some catastrophic errors in my interpretation of your interpretation, but the assumptions in the macroeconomics here seem to be a poster-child for the appalling use of data. Tax comes out, redistribution goes back in, average person pays and receives average and negating amounts, thus it's all hunky-dory? If this is macroeconomics, I'm terrified as an educated outsider.

Garett Jones writes:

@Richard:

Actually, the usual "chapter 1" assumption in neoclassical models (Thomas Sargent's dynamic macro textbooks, for instance) is that government spending is pure waste. Potatoes thrown in the river.

But Prescott's tale is that if you accept the welfare state's supporters optimistic claims about government spending, then that makes government spending even worse for the job market.

Hoist on their own petard, as Hamlet might say.

Kevin Dick writes:

OK. That... is... awesome. A big government either destroy value or destroys the incentive to work.

Also, this inspired the thought that a wisely spending government is isomorphic to the Diner's Dilemma. Is that right?

david writes:

It's "even worse" in the sense that efficient transfers make the recipients want to work less, since they have more after-transfer income. It's intuitive if you think of it that way.

Garett Jones writes:

@Kevin:

Indeed. Google isn't helping me find Russ's classic op-ed on the Diner's Dilemma---I'd appreciate it if someone could put up a link to that in the comments....

david writes:

Note that, e.g., in a purely efficient pure transfer tax, with zero MTR on the incentive to work, we would still likewise expect labour force participation to fall for exactly the same effect outlined here.

This is because Prescott is not comparing the MTR generated by the tax hike with no tax hike, but the MTR generated by the benefit hike with no benefit hike. And there is no presumption here that the benefits are conditional on not working; the result springs entirely from the argument that people would work less if they found it easier not to starve.

john hare writes:

So if more money is stolen/taxed from me, I'll have incentive to work harder? If it's all the same to you, I'd rather work less for the same reward, or just as much and have some investment for my future.

Chris H writes:

Now this is a fun little argument Prescott's produced here!

The more assumptions of the other side you can accept and still prove their conclusions invalid the better. Now I happen to think the boondoggle assumption is likely closer to the truth, but this means that if welfare-state proponents are right that the spending is helpful, then this spending ultimately undermines their own program! The more effective the hand-outs are the more the taxes needed to support them will drive people into non-production this then weakens your tax revenues meaning you either have to accept your programs becoming increasingly ineffective or raise taxes which starts the cycle all over again.

But if I may I'd like to take this even further. Now of course, maybe they'll argue this is why a deficit is good because funding from loans avoids the tax substitution problem. But even assuming that at no point will taxes have to be increased to pay for increasing deficits we still have a problem. Taking loans means those loanable funds won't be available to private interests who would use it not to pay for consumption purposes as hand-outs do, but for investments that increase the capital stock of the economy. If capital investments are thus discouraged, workers can't increase their productivity. If that can't happen the only way the economy can grow is simply population increase, but per capita growth and economic progress is stalled. This then prevents the introduction of more resources into the economy, resources often needed to support these welfare schemes (where are the rare earth minerals needed to make a lot of modern medical equipment going to come from if the loans needed to start up mines are all taken up by the welfare state?).

Now the last means to try and pay for this welfare state is inflating the money supply to artificially increase the amount of loanable funds. But inflation hurts those on fixed incomes, aka those receiving welfare from the state. Thus you are left once again with the trade off of making your programs gradually more inefficient, or indexing the benefits to inflation. If you do the latter, the entire point of the inflation is defeated as the loans needed to pay for the programs will therefore expand with inflation. Thus the expansion of loanable funds will be matched by the expansion in costs of the program and you're back to the "taking money from investment" problem of the previous paragraph.

The end result? State welfare programs must always be either ineffective boondoggles (in which case why would any honest person support them?) or effective programs that gradually destroy the proper functioning of the economy they exist in.

david writes:

I will point out again (and probably fruitlessly), that the dynamic Prescott has identified is elegant, but not in any politically applicable way.

It's the same as if some huge labour-saving technology were discovered tomorrow: of course labour force participation falls when there is less waste of outputs of labour, holding income effect nil. It's the same whether the waste is reduced via technology or via hypothetically effective programs.

david writes:

(also, anytime one sees suggestions for a Hayek or Friedman-style replacement of the existing welfare system with guaranteed fixed lumps of payouts, this is exactly the empirical problem. Labour force participation near the bottom end falls too much.)

Max writes:
Indeed. Google isn't helping me find Russ's classic op-ed on the Diner's Dilemma---I'd appreciate it if someone could put up a link to that in the comments....

http://www.invisibleheart.com/Iheart/PolicySirloin.html

Richard Manns writes:

@Garett Jones

Thank you for pointing out my catastrophic error!

I read 'employment' as 'unemployment' for some unknown reason!

As you can imagine, I was massively confused as to why the result was even publishable, being filled with such mad assumptions.

You've enlightened me. Thank you.

KevinC writes:

Garret Jones -

Here is a link to Russ Roberts on the Diner's Dilemma. It's also published as an essay in The Libertarian Reader.

Best,

Kevin

Methinks writes:

Well played, sir. well played!

On the tangential topic of government waste, I toss this article by James Payne onto the table. Perhaps around $6 Trillion is wasted annually; 65% of it on the act of collecting taxes alone.

Charlie writes:

The argument doesn't work very well for top marginal rates. Aren't economists on both sides for broader bases? Read Saez on optimal taxation.

Tom West writes:

The Diner article was interesting, and brought up a good point about behaviour when someone else is footing the bill. But I feel it weakened the case by trying to generalize beyond the base point, especially with its two final questions:

How many Americans other than farmers benefit from the farm subsidy programs?

I believe the justification for the program (whether it's valid or not) is food security. It doesn't take a genius to see that in times of great stress, market functions between countries break down and a country being unable to feed oneself would be catastrophic. Of course, whether such programs provide the promised security or not is a different question, but the presumed benefit is not "their times at the trough", but insurance of the sort that we as a whole have come to expect our government to provide.

How many Americans other than train riders derive benefit from the Amtrak subsidy?

Presumably, all benefit from those not competing for scarce road resources and the smaller amount of pollution. (And again, whether it accomplishes those goals is a different matter.)

Neither are good examples of his original point.

Kevin Dick writes:

Thanks Garret and KevinC.

I was actually thinking of the Diner's Dilemma in the sense of a disincentive to work (which I now realize is wrong--that's sort of a converse).

So there are actually _three_ problems with progressive government programs that make us worse off:

(1) The ineffective programs are directly destroying value.

(2) Effective government programs reduce the incentive to work, decreasing aggregate income.

(3) Effective government programs increase consumption of those programs' outputs beyond the economically efficient level.

MingoV writes:

Government, regardless of its efficiency, never takes money out of each taxpayer's right pocket and puts the same amount into each taxpayer's left pocket. There always is redistribution. The redistribution in the USA today (and in the foreseeable future) takes tax money from the right pockets of middle- and upper-income workers and puts much of it into the left pockets of the no- or low-income workers and the (mostly non-working) elderly.

Thus, for tax increases on middle- and upper-income working people, the income effect remains, and Dr. Prescott's conclusions are incorrect.

Methinks writes:

but the presumed benefit is not "their times at the trough", but insurance of the sort that we as a whole have come to expect our government to provide.

Let's just buy this flimsy argument for one second (we have plenty of ability to produce without subsidies, btw). That doesn't make it okay. How much are we paying for this insurance?

If we're overpaying (and it's pretty dang clear we are), then it's not worth it. By definition.

Same thing for Amtrak.

honeyoak writes:

While I love the logic of Prescott's paper, I remember running the regressions myself and finding no significant results with an expanded data-set.

Carl writes:
"I believe the justification for the program (whether it's valid or not) is food security. It doesn't take a genius to see that in times of great stress, market functions between countries break down and a country being unable to feed oneself would be catastrophic. Of course, whether such programs provide the promised security or not is a different question, but the presumed benefit is not "their times at the trough", but insurance of the sort that we as a whole have come to expect our government to provide."

Are you explaining the government's P.R or are you actually saying that "food security" is a major problem for the U.S which must be dealt with by subsidies? Trade is always going on, in periods of "great stress" or otherwise, and the population of the U.S can feed itself just fine without government subsidies to farmers.

I have no idea what "we as a whole" means, let alone what the expectations of this weird abstraction could be. Who do you speak for when you say "our government"? Just yourself? Or yourself plus 311 million others? Please stop abusing pronouns, it's giving me a headache.

CC writes:

Didn't Mankiw write about this back in 2007?

Isaac writes:

Stuff like this goes through my head all day. Thank god there are articulate people like you that bring it to paper and give it credibility.

Tom West writes:

Are you explaining the government's P.R

I was explaining the justification that is used by governments around the world for subsidizing farmers.

Personally, I'm not certain that in America's case, the subsidies are either necessary or help accomplish the goal, but the concept of food security is not entirely stupid, as we witnessed many countries halting food exports when they encountered difficulties.

In the event of severe problems (like a devastating wheat disease), I could easily see the US and Canada barring exports, leaving countries currently dependent upon North American wheat for basic food needs starving regardless of how much money they would be willing to pay.

As for confusing pronouns, I think I can safely say that the majority of voters on both right and left hold governments responsible when the government fails to prepare adequately for large-scale disasters.

Chris H writes:

Tom West writes:

Personally, I'm not certain that in America's case, the subsidies are either necessary or help accomplish the goal, but the concept of food security is not entirely stupid, as we witnessed many countries halting food exports when they encountered difficulties.

Arguing in favor of "food security" policies on the basis that "food security" policies are used seems specious to me. But let's think for a second what happens if a country bans food exports.

From the farmer's perspective, what has been done is that they have lost a lucrative market and can no longer sell their products at the price they could once get. Sounds like a win for the country's consumers right? Except that what this price drop has done is disincentivized food production. Farmers will thus produce less food than before, meaning that there will not be an impact on domestic prices. Thus you are not in fact helping the domestic consumers and you're harming foreign consumers and the farmers. If farmers are not selling to domestic consumers in the case of a famine, this is not because they are greedy but because the domestic society cannot actually economically support the level of food production needed to support it's population. In the modern world, the most likely reason for this to be the case is because of government policies that impoverish the people.

Why then do governments engage in anti-food exporting policies if it doesn't actually help the people? It's a panacea and an attempt to keep blame away from themselves. People don't actually know the economics which would tell them what course to take to reduce the food crisis and thus latch onto things which make some intuitive sense. If you don't ship food overseas you'll have more at home right? Governments, wanting to seem like they're helping the crisis and avoiding the wrath of an angry public at their "inaction" follow through on this incorrect conclusion the people have come to. The result is not food security, but actually less prosperity for the society which in the long run is the true cause of food security. A prosperous society has no real fear of running out of food, but a society which impoverishes itself for fear of not having food will have a higher likelihood of running into problems.

Charlie writes:

"But let's think for a second what happens if a country bans food exports."

I'm pretty sure Tom was talking about the case that actually happens in the U.S., which is subsidizing food production and export. The U.S. floods itself and the rest of the world with under priced corn, grains and other staples with lots of subsidies.

If you point this out, people will often give the counter argument that without these subsidies, the U.S. might become a net food importer through a vast and miraculous network only possible through world trade. But during WW III this miraculous network may break down, when our enemies aerial drones destroy our supply chains.

This is the argument. It sounds like neither Tom or I think it carries the day. But it isn't a stupid or unreasonable counterargument, and must be dealt with.

Lastly, this is totally off topic, so stop trolling Tom about Ag Subsidies.

Tom West writes:

Sorry for derailing the topic - this is my fault.

Let me just state that I think North American agricultural subsidies are almost certainly unnecessary and I fully understand that halting exports is a bad idea. However, we saw examples of countries doing exactly this only last year, so assuming that your trading partners will remain intelligent in the face of crisis is not necessarily a good bet :-).

andy writes:

This reminds me of a different idea - with lot of corruption government spending is wrongly counted as consumption instead of transfer. Therefore you should see higher fiscal multiplier (in the short run) in countries with high corruption...

Ritwik writes:

My first thoughts when I first read the Prescott paper a couple of years ago were - wow, so earlier Europe used to work as much as the US for 3/4th the output and now it works 3/4th the time on 3/4th the output! What explains this superior productivity growth in an era of reduced work effort?

Jonathan writes:

What about when government spends tax money on wise investments rather than redistribution? Consider a government that invests in valuable projects that are too big for a single private entity to undertake - like building roads and bridges, funding basic scientific research, or educating our nation's young. If the government spent tax money well, then that would have a positive income effect, because society as a whole is enriched. A wise government is an admittedly high bar to clear, but the theory still has some hope.

Seth writes:

Sounds like the free rider problem.

Maybe Romney was onto something?

K writes:

Hey, cool! So for rich people, who effectively get nothing from the redistribution, the effect of taxes on work is ambiguous. So much for Galt's Gulch!

Sonia Conly writes:

Not such a high bar to hurdle when administrative costs are considered. Consider Medicare. The direct administrative costs (excess of premium over payment) of Medicare are a few percent. The direct administrative costs of employer plans are in the range of 10-15% and the admin costs of non-group insurance in excess of 20%. Similarly the administrative costs of Social Security are far less than the administrative costs of e.g. 401 k plans.

Personally I place a high value on the transfers provided to less fortunate persons. My total welfare is increased knowing that others are provided for. Sonia

Ray Lopez writes:

@ cc: "Didn't Mankiw write about this back in 2007?" http://gregmankiw.blogspot.gr/2007/04/do-income-effects-matter-for-tax-policy.html yes he did. But I swear I've seen this in a textbook somewhere too...

Methinks writes:

So much for Galt's Gulch!

Yeah....not so fast. Very high earners tend to consume a small portion of their income, so they can significantly drop income without forgoing consumption. They'll still feel rich. I suspect the substitution effect would be pretty significant for them.

They also have a lot more freedom to mess around with how they receive their income and will divert a lot more time to figuring out how to avoid paying taxes.

There's more than one way to "go Galt".

The 1986 tax reform dropped the top marginal tax rate to 28% from the 70's. The tax revenue from that group skyrocketed as a result. Muni bonds and complicated tax sheltered structures just weren't worth it anymore.

Chris H writes:

Oh! Sorry Tom I misunderstood your point!

Methinks writes:

Personally I place a high value on the transfers provided to less fortunate persons. My total welfare is increased knowing that others are provided for. Sonia

That's great. Have you ever checked whether these people are actually being provided for or does your concern end with your support of a government plan that confiscates from producers?

For every dollar taxed away, the needy that you claim to feel so strongly about get something like $0.02. Most of it goes to running the bureaucracy. Or maybe it's bureaucrats you were concerned about in the first place. Hard to tell. And when the needy do receive whatever is left over after the bureaucracy has taken the lion's share, they get don't get those scraps in cash to decide for themselves how to spend. No, we equate poverty with stupidity. They get....government cheese; which they promptly march to the street corner and monetize for $6 to a passing motorist.

I used to live in that appalling mess you think passes for "care of the less fortunate". If that's what your compassion looks like, they're better off without it.

Of course, you might consider that if actually taking care of the less fortunate so increases your welfare, you could just give to charity. Charities compete for my dollars and I always choose the one where 98% of each dollar goes to the intended group, not the bureaucracy. Charity self-taxation and most Americans are happy to do it.

Oh, and since when was Medicare ever restricted to the "less fortunate". It's just one of those government schemes which crowded out a private market in health insurance for seniors.

Coding Monkey writes:

I find it kind of amusing when people act like the government cannot possibly innovate. Like there is some sort of basic difference between people working for the government and those working in private companies. Do you imagine the only people who can innovate are those whose incomes are directly linked to profits of those innovations? I hope this is not true since that would eliminate the vast majority of workers as having any chance of being an innovator. This is a great thought experiment but its ties to reality are tenuous at best.

K writes:

Methinks,

"Yeah....not so fast. Very high earners tend to consume a small portion of their income, so they can significantly drop income without forgoing consumption. They'll still feel rich."

But if they are working it's in order to satisfy future consumption plans the realization of which now require more work than ever. The rich also have hopes and dreams. And *lots* of them do have debt and very high consumption rates to which they have become accustomed. *Especially* the ones who are actually working.

I vote for "ambiguous".

David C writes:

I thought this was brilliant, then after some considerable thought realized it had little applicability to the real world. Because governments are able to take on such massive debt loads, government spending levels aren't correlated with taxation levels. Yes, theoretically, revenue should have to eventually equal spending, but this process takes decades, if not centuries. But then time preferences have to be taken into account.

In terms of immediate effects, lower tax rates tend to increase government spending, and higher tax rates tend to decrease government spending. Source: http://elsa.berkeley.edu/~dromer/papers/draft509.pdf

So....
If higher taxes decrease the number of valuable government programs, then a tax hike causes a big increase in employment.

Methinks writes:

But if they are working it's in order to satisfy future consumption plans the realization of which now require more work than ever.

Of course. There are several things they can do, but the most obvious is to re-evaluate the value of that future consumption. When the price of something changes, don't you re-evaluate your consumption of it?

The rich also have hopes and dreams.

This I just find strange because hopes and dreams are not unique to the rich. Still, hopes and dreams are not priceless.

And *lots* of them do have debt and very high consumption rates to which they have become accustomed.

Well, having debt doesn't mean you have unmanageable debt and high earners tend to consume a fraction of what they earn, unlike the middle class which tends to consume everything and save virtually nothing. So, unless your hypothetical high earner is unusually profligate, he'll not have to drop his consumption if his income drops. However, as I've said before, "the rich" also have a lot more flexibility in how they take their income and have the most flexibility to adjust it to lower their tax burden. Frankly, they also have more flexibility in how they consume. Yes, "how" not "how much". Look for more pre-tax consumption in the form of company cars, planes, apartments, houses, and vacations. I know a doctor who holds seminars in exotic locations in large part so that his family vacations are an expense line item in his practice. Voila! Pre-tax consumption.

I vote for "ambiguous".

My hypothesis is that the substitution effect is bigger for the wealthy. More importantly, my hypothesis is that the substitution effect grows stronger as income rises. You can live just as well on $2 million per year as you can on $4 million but with A LOT less effort and risk. Also, I hypothesize that while you might be right that the rich might work just as much, what they work on will change. Much more time and energy will be diverted from earning a taxable income to sheltering income and consumption. If the goal is to collect more tax revenue from them, you'll see the opposite.

That "the rich" manage their tax burden is pretty obvious. After the 1986 tax reform (when tax rates were cut for all brackets), the tax revenue from the top tax bracket increased dramatically, more than offsetting the decrease in revenue from the middle brackets (who tend to spend everything they earn). It's the middle class that has fewer options, not "the rich".

And I caution that "the rich" are rich in ways that aren't easily accounted for on a ledger. These are people with the most desired skills, the better ability to raise money for new businesses anywhere in the world and the most options. That's not stuff you can tax away from them, but it is the stuff most desired anywhere in the world.

Ohio libertarian writes:

Think about the "round trips" most Federal tax revenue makes between your pocket, DC and back again for:

1) Education
2) Transportation
3) Energy
4) Housing (HUD)


A dollar to DC, skim off $.xx for operating the relevant bureaucracy, then $1- $.xx back into "somebody's" local economy.

Explain again how a "positive multiplier" is generated?

And I haven't touched on agri-industrial subsidies and entitlements...

Methinks writes:

I find it kind of amusing when people act like the government cannot possibly innovate.

Nobody worth listening to says that. Even advocates of government recognize that the expectancy is negative. As long as the expectancy is negative, government is waste at worst and break-even at best.

Like there is some sort of basic difference between people working for the government and those working in private companies.

There is. It's called "incentives". First of all, nobody does government work in order to innovate. What incentive to do they have to innovate for the profit of the government where they can't personally profit from innovation. Even if they do want to innovate, such a thing is virtually impossible in a government bureaucracy. The only thing government ends up innovating is ways to rob you.

Methinks writes:

to correct a mistake - if the expectancy is negative, we expect to lose on every investment made by government. Break-even is our best hope. But, if you're hoping and praying, you're in a bad trade.

Comments for this entry have been closed
Return to top