Macro question

by Russ Roberts on January 6, 2009

in Stimulus

Here are three $750 billion stimulus plans. All three are financed by borrowing:

1. Give every American a "tax rebate" of $2500.
2. Hire 10,000,000 Americans and pay them $75,000 to dig a ditch for six months and spend the next six months filling it back in.
3. Hire 10,000,000 Americans and pay them $75,000 to build bridges and sewers and landscape parks and highways for a year.

Do any of these stimulate the economy? If yes, is there a difference between which does a better job at stimulating the economy? You may continue your answer on the back of the page.

Extra credit: Combine 1 and 3. Give every American $1250 and hire 5,000,000 Americans to build bridges etc. Same effects? Different? Why?

Extra extra credit: Would macroeconomists agree on the answers to these questions? Why not? What empirical evidence might be used to make a case for one side or the other?

Be Sociable, Share!



72 comments    Share Share    Print    Email


scott clark January 6, 2009 at 10:04 am

Ha! Is it too late to drop this class?

Randy January 6, 2009 at 10:07 am

You can give me an F if you want, but I'm going to BS my way through the question…

Its a trick question. Its based on the assumption that there is such a thing as "the economy". There isn't. In each and every option, some people will gain and some people will lose. And in each and every option, the government will be able to point out the winners and say, "Look there, we did good". They will, of course, ignore the losers. My favorite, if I have to choose, would be option 1, because that will stimulate my economy.

Doug Ransom January 6, 2009 at 10:21 am

1. will be neutral. I recall Ricardian equivalance being discussed on this blog?
2. destorys wealth.
3. At least this option builds infrastructure the nation needs and increases the nations wealth and productivity- if we need new infrastructure it makes sense to do infrastructure spending counter to economic cycles, so the economy is stabilized somewhat and infrasturucture spending is not competing for same resources as private sector expansion/emploment.

Rex Pjesky January 6, 2009 at 10:26 am

Short answer….all of the policies fail to stimulate the economy. Instead, the three policies simply shuffle resources around, redistributing them to the ditch diggers, highway workers, or current taxpayers.

Macroeconomists will disagree with one another on this point based on what they believe about the multiplier. Economists who believe in the multiplier within what has become known as a Keynesian framework will think that ANY of these proposals will stimulate the economy. Economists who don't believe in the multiplier will not.

AisA January 6, 2009 at 10:39 am

I agree with Randy and Rex. Obama's stimulus plan will be every bit as successful as was Hoover's, Roosevelt's and Bush's.

The government does not have the power to create demand or "purchasing power" — it has only the power to shift it around from one sector or person to another.

Zachary Kurtz January 6, 2009 at 10:48 am

Option 3 may improve longterm efficiency by affecting the productivity of the economy. Ultimately, though, infrastructure costs will remain high if technological innovations aren't made to reduce repair costs, etc.

Option 1 would create general inflation because it would drive prices up equal to the money received.

Option 2 is akin to burning money to keep people busy. It will also generate inflation, which will punish those who aren't working.

Zachary Kurtz January 6, 2009 at 10:51 am

Option 3 may improve longterm efficiency by affecting the productivity of the economy. Ultimately, though, infrastructure costs will remain high if technological innovations aren't made to reduce repair costs, etc.

Option 1 would create general inflation because it would drive prices up equal to the money received.

Option 2 is akin to burning money to keep people busy. It will also generate inflation, which will punish those who aren't working.

Marina martin January 6, 2009 at 10:51 am

None stimulate the economy (unless stimulation can be negative?). Number three seems worst, because all the private infrastructure firms are put out of business in addition to every American (now and future) being saddled with additional debt that damages every industry.

Roger January 6, 2009 at 10:56 am

1. Give every American a "tax rebate" of $2500.
This mostly depends on how you think the people will spend the money they are given. If they mostly understand and believe in Ricardian equivalence then the money will simply be saved to pay off future taxes, with no net benefit. However, and I think this is the more accurate view, if a significant portion of the rebate is spent, then it certainly can benefit the economy somewhat depending on how large the multiplier effects are, though at the risk of creating future problems.

However, this is only if people believe that the government will not inflate in order to pay off debt. This infusion of cash will also probably create inflation, there could be some additional benefit since nominal wages are sticky downwards and this would allow firms to cut real wages without affecting morale or losing talent.

2. Hire 10,000,000 Americans and pay them $75,000 to dig a ditch for six months and spend the next six months filling it back in.
This is the worst idea by far. Although this fulfills much the same basic goal as the above proposal, injecting money into the economy to stimulate demand, it comes with many unnecessary downsides. A huge subsidy is created for unproductive jobs, inducing people to lose leisure time without creating anything of worth and decreasing overall utility. In addition, the subsidy affects a much smaller amount of the population, redirecting wealth towards those chosen. This also adds the problem of politics, where jobs are given not to the “neediest” but to the most politically-connected. And finally, although this effect will be very different depending on the relative pay and amount of the theoretical jobs, this will sap productive employment from the private sector and create additional competition for labor, making it harder for businesses to attract and keep talent and delaying recovery.

3. Hire 10,000,000 Americans and pay them $75,000 to build bridges and sewers and landscape parks and highways for a year.
This has all the same problems and benefits as the above except that instead of doing worthless things and creating no worth, these jobs are doing less useful things but at least creating something.

Rex Pjesky January 6, 2009 at 10:56 am

Option 3 no doubt looks attractive to many, but remember the opportunity cost. The resources used in infrastructure spending is taken from some other use that may be more productive.

I firmly believe that some infrastructure spending might be of a long term benefit, but I don't trust the government to do it properly.

And, like Tyler Cowen thinks, it does not (and should not) be done only in the name of stimulus. If it is a good idea when the economy is bad, it would have been a good idea when the economy was good.

Brian January 6, 2009 at 11:01 am

"Do any of these stimulate the economy?"

No. In all three cases, the impact of this redistribution would only serve to partially offset the impact of confiscating the money in the first place.

"Combine 1 and 3. Give every American $1250 and hire 5,000,000 Americans to build bridges etc. Same effects? Different? Why?"

Same effect as the answer to the first. Any perceived stimulation is likely to be lower with a combination of efforts, given the increased overhead of doing both.

"Would macroeconomists agree on the answers to these questions? Why not? What empirical evidence might be used to make a case for one side or the other?"

Most macroeconomists I've read would disagree. I don't know why, but I'm sure they all have their various reasons. For empirical evidence, consider whether any other example of theft has resulted in stimulating an economy. Any macroeconomists that disagree are welcome to hang out in dark alleys with wads of cash hanging out of the pockets to help stimulate the economy. Make sure Krugman doesn't forget that $1.4 mill Nobel cash award.

SheetWise January 6, 2009 at 11:04 am

Option 1 is giving Peters money to Paul. No change.

Option 2 is giving Peters money to Paul while punishing Paul. A lot of wasted effort.

Option 3 is creating wealth — but is very inefficient.

Bret January 6, 2009 at 11:04 am

1: Depends.

2: Depends.

3: Depends.

1 & 3: Depends. Same effect.

Extra extra credit: Wouldn't agree. Economists never agree on anything. Empirical evidence? We're talking economics here – most economists seem to abhor empirical evidence.

Bill Nelson January 6, 2009 at 11:16 am

1. This sounds like a $2500 loan to every American — from every American. Out of my right pocket and into my left. Plus transaction costs. Sounds like a bad deal.

2. This shifts property from some people to some other people, and it's pernicious because the new ditch diggers will no longer be producing anything of value. Furthermore, in the long run, the tax burden will discourage non-ditch diggers from producing, and wealth further diminishes.

So far, I prefer #1 because at least people are not taken away from productive jobs.

3. This is very similar to #2 in that much of the money will be directed to things that no one wants. But at least some value is created by building bridges. Perhaps those resources could have been put to better use, but at least it might not be a total loss.

I don't know what the effects would be on economy "stimulation" as I have no idea what the phrase "stimulate the economy" even means.

It sounds pretty distasteful, though.

Bill January 6, 2009 at 11:18 am

It is assumed that all three are financed by borrowing. I recall a clever fellow writing that these are analogous to dipping water out of one end of a pool and pouring it in the other end.

Rob Dawg January 6, 2009 at 11:39 am

Option 3 has the possibility of being even more anti-stimulus than the first two. Infrastructure comes with carrying costs. In the case of infrastructure like rail transit projects the ongoing operating subsidies will continue to drain resources long after the project is completed.

Chris January 6, 2009 at 11:48 am

#2 and #3 are anti-stimulus. If the idea of "Stimulus" is to help the economy do better in the future, then having a bunch of people go out learn to be the equivalent of wig-powderers (a trade which has no future) is the wrong way to go about it.

Among all these, I would prefer #1 because, apart from the marginal governmental waste in printing checks, it has no effect.

Dave Peterson January 6, 2009 at 11:53 am

Don't you need to know what rate government is borrowing at? Theoretically you could build or improve infrastructure that could provide a return greater than what you borrowed it at. I know if I spent less time stuck in traffic around DC I'd be richer (and probably even richer to the point that I'd save more in gasoline and car maintenance than my share of the added tax burden of those jobs). But I also know that these stimuli don't work that neatly and efficiently in practice.

Martin Brock January 6, 2009 at 12:03 pm

Public goods contribute to many productive organizations, but no organization can effectively own one without creating excessively restrictive monopoly; therefore, production and organization of the goods for profit is not possible or counterproductive. These goods certainly exist in theory and in practice. Roads are classical examples. Knowledge from research is too. 3. potentially creates public goods.

The problem with public goods is that politicians can and will call anything they like a "public good" whether or not it contributes much value to any productive organization. When aggregate demand is insufficient to employ idle resources, taxing or borrowing to creating public goods can be useful in principle, but other approaches are possible. A monetary authority may extend credit to organize idle factors. The typical object to this approach is that "no one is willing to borrow", but "no one" in this formulation typically describes some existing corporatist structure already laden with debt.

Let me incorporate, and I'll try to organize a few idle factors profitably in a year or two. Expect me to gamble my children's food, clothing and shelter, or even my minimal support in old age, on the organization, and I won't. People like me are never in short supply. Only the credit can be in short supply. Extending it without putting the borrower much at risk can be counterproductive, but this approach is probably more productive than extending the credit to politicians creating "public goods".

Randy January 6, 2009 at 12:06 pm


Just an aside, but I recently had the opportunity of driving on the northern part of the beltway. Beautiful roads. I was thinking while creeping along in the bumper to bumper traffic that this probably looked really good on paper.

Market Urbanism January 6, 2009 at 12:08 pm

Could #1 stimulate simply because the government's cost of borrowing is cheaper than the individual taxpayer's opportunity cost of capital? (aside from all the distortive effects) Thus, if individuals invest their tax savings in private enterprises, growth will be faster than otherwise, as long as the government's credit rating is not reduced in the process. Or is there more to it?

Adam Malone January 6, 2009 at 12:32 pm

Short answer: Governments do not have the power to stimulate economy. The have the ability to decrease their negative impacts on the economy, but apart from enforcement of contracts and rule of law, government's ability to stimulate the economy is non existent.

As stated above, they simply shuffle money from one group to another. What many do not consider when discussing the government's efforts to redistribute wealth is that it is by definition an inefficient mechanism. When the government takes $100 from me to give to someone whose need is "greater" they do not give $100 worth of assistance to the person. The IRS has to account for the money, the treasury department has to send it out, etc.

Every step of the way a small portion of that money is eaten away due to administrative costs.

So, if the government takes my $100 (or get's a loan and makes my child pay for it later on) and goes to build a road with it I will not only be forced to spend my money on something unnecessary to me, I will actually get LESS than $100 worth of goods (be they bridges, roads or widgets). Talk about a raw deal.

John Dewey January 6, 2009 at 12:42 pm

Way back in January, Russ suggested that Americans would be hesitant to spend a rebate check. He offerred that Americans might save the rebate, knowing they would eventually have to pay it back in higher taxes. I was not convinced:

"I have great respect for Professor Roberts also, but I disagree with his suggestion that families will not spend the windfall. Many of my lower middle income relatives will spend their share before it ever gets to them.

Since then I've become more convinced. Most Americans pay little if any income taxes. If they even consider that rebates must be repaid – my relatives have no clue about that – they likely believe that only the rich will be burdened with repayment.

Are rebates a simple redistribution? It seems to me we will be increasing demand in period 1 and reducing demand in periods 2, 3, 4, etc whenever taxes must be raised. I'm not sure that smoothing economic activity should be a function of government, but isn't that what the rebates accomplish?

Oil Shock January 6, 2009 at 12:46 pm

If dissipating wealth through options 1, 2 & 3 is what stimulated the economy, Soviet Union will still be around and will be the lone super power. North Korea would be a paradise. Cuba will build walls around it to stop Floridian refugees.

All the poor socialist nations in Africa, Central America or Asia can become so unbelievably wealthy by spending themselves to prosperity.

Oh, What a wonderful fantasy!

Mike January 6, 2009 at 12:47 pm

First let me put aside any moral concerns about whether or not the government OUGHT to borrow to stimulate. Let stimulative mean productivity higher than would have otherwise occurred. From a purely theoretical standpoint of CAN they stimulate, I think the answer is a qualified yes.

Option 2 can never be stimulative because holes full of dirt do not earn a rate of return above the government's cost of capital.

For option 3 to be stimulative, the following we need to be true: 1. There exist projects that the government could undertake that would earn returns in excess of their borrowing costs. (even accounting for cronyism, etc.) 2. Private borrowers will not undertake these profitable projects due to IRRATIONAL fear about their future returns or IRRATIONALLY high costs of capital from private lenders.
Under these conditions, government borrowing could stimulate real productivity gains i.e. not just create future inflation or tax increases.

Option 3 could stimulate in an extremely unlikely or roundabout way. 1) Gov't borrows from entities unwilling to lend to ANYONE else 2)People use check to pay down debt 3) Private lenders who were previously fully constrained by their balance sheets use the newly created excess lending capacity to fund capital projects for which there was demand for funds but no previous private supply.

Doubt this could occur.

I believe there are two major macroeconomic disagreements: 1) The degree to which you believe that liquidity traps are possible AND that the Government could earn its cost of capital when no one else is able or willing to try. 2) The degree to which this type of behavior is morally desirable or acceptable or even at all desirable from a more long run moral hazard type view of the future behavior of all economic actors.

Oil Shock January 6, 2009 at 12:52 pm

Bret says….

most economists seem to abhor empirical evidence.

What are you smoking? All mainstream economists can quote statistics to support their own selfish agenda. It is easy to take up position on any economic matter, for or against, based on statistics. Only the HETERODOX Austrian school insists that you need theory to interpret data, and not develop theory to suit your data.

Xmas January 6, 2009 at 12:52 pm

All three are pointless if this isn't a shift of existing government spending. If this is "spending" on top while keeping other spending the same, then this is really poor.

Though, building infrastructure would be a nice thing for the government to ACTUALLY spend money on.

Shawn January 6, 2009 at 1:22 pm

1,2,3: stimulus FAIL.
(see: unseen, a la bastiat)

Bill January 6, 2009 at 1:57 pm

The notion that peoples' saving a rebate check dampens the stimulative effect of such checks seems odd to me. Saving is not necessarily the same as hoarding (e.g., placing the check under a mattress or otherwise removing the purchasing power from circulation.) If I save my check by depositing it in a financial institution, and that institution, in turn, lends the money to someone else who makes a purchase, how is that any less stimulative than if I spend the money?

ScooterC January 6, 2009 at 2:09 pm

I don't think #3 is a $750 billion option. The cost of materials for bridges, sewers, etc. is really high. I'd think you'd spend twice as much on materials as labor in a project like that. There's been quite a bit of discussion of "the government's cost of capital", but what about the cost of materials? What country's economy would we really be stimulating by buying all that cement and steel?

Sam Grove January 7, 2009 at 1:58 am

The purpose of government stimulus is to redistribute stimulus in such a way as to produce visible effects while the unseen effects are ignored or blamed on the usual suspects.

There is always stimulus. The profit motive stimulates people to engage in productive behavior. The reduction in this area will be the unseen effects .

Ike January 7, 2009 at 12:46 pm

No one has mentioned that by hiring 10,000,000 people at taxpayer expense, we are also artificially jacking up prevailing wages by removing competition for jobs. This added expense will further harm overall productivity, and lead to cloudy market signals on top of the inflation caused by borrowing.

Martin Brock January 7, 2009 at 1:02 pm

The cost of materials for bridges, sewers, etc. is really high. I'd think you'd spend twice as much on materials as labor in a project like that.

Where do the materials come from? Someone mines the limestone and mixes the cement, and people manufacture machinery for these purposes. Labor is most of the cost of everything except the rawest of raw materials.

Congress could also pay people to build more houses, but we arguably have too many houses already. "Infrastructure" and other public goods are valuable only insofar as they add sufficient value to other productive endeavors, but with no market to calculate this value, it can be difficult to gauge.

Suppose we have a lottery. Winners receive anywhere from a million to a hundred million dollars, but they may not simply consume it. They must invest it, and by "invest", I don't mean buying "securities".

Winners incorporate. Congress owns 90% of each corporation's shares, and the winning founder owns the other 10%. The corporations purchase means of production, including labor, and produce goods for sale. A corporation may pay its founding winner a salary limited by law. After five years, if a corporation has a market value, Congress sells its shares; otherwise, it dissolves.

Taxpayers might lose more than it gains this way, but at least we'd know what we lost and how we lost it, and we'd also know what we gained and how we gained it. With "infrastructure", we only know that the bridge is out there and crossed occasionally. Since no one pays to cross, we don't know much about the value of the crossings.

Brian January 7, 2009 at 4:30 pm

I wish to modify my original answer to the first question. There is a way the first option could stimulate the economy: write the legislation for this $2500 "tax rebate" so that it's not distributed as checks, cash, or debit cards, but as gift cards that can only be used to purchase pitchforks, torches, rope, and transportation to DC. Perhaps additional stipends could be included for tar, feathers, cattle prods, etc.; firearms might be a little too controversial.

This could be like the angry mob version of food stamps.

This also modifies my answer to the extra credit questions. Combining 1 and 3 would provide less money for the angry mob, but the infrastructure improvements may make it easier for everyone to get to DC without too much traffic.

I would still expect the macroeconomists to disagree, but you might be able to get a few more on board given their penchant for creative destruction. In that regard, it's somewhat similar to the second option of digging a bunch of holes which most of the macroeconomists seem to favor.

Bill Woolsey January 8, 2009 at 10:03 am

So far, the answers are quite poor.

Generally, "stimulate" refers to increased spending (nominal) on final good and services.

Because all three programs are financed by borrowing, the question is whether or not increased government borrowing reduces the demand for money. That is, if it somehow causes people to hold less money than they otherwise would. If people hold new government bonds in place of existing money balances, total spending might rise.

If there is no offsetting change in the quantity of money, reduced money demand creates a surplus of money and increases total spending in the economy. It "stimulates" the economy.

If the price level is currenty at equilibrium, the result is monetary disequilibrium, that will cleared up in the long run by a higher price level.

If the price level is currently above equilbirium, then the policies will move the equilibrium price level towards its existing level, moving the economy towards equilibrium. In this situation, the incease in government bonds and reduced demand for money can increase total production and employment in the short run.

There are two issues that would divide economists. One is the degree to which government bonds can substitute for money. This was traditionally debated in the context of the interest elasticity of the demand for money. If the context of the question is now, then how does the unusual situation of the interest rate on short government bonds being only a few basis points impact the situation.

A second element of division among economists is whether or not the price level is always at equilibrium, or is it possible for it to be above equilibrium in the short run. This discussion has been traditionally in the context of the short run phillips curve.

All three programs will likely impact the allocation of resources.

Number 1. will likely shift resources from investment to consumption, reducing real income and welfare in the long run. Other things being equal, the lower income implies a higher price level in the long run. If everyone is a perfectly informed rational mazimizer, this won't happen.

Number 2. will reduce real income and welfare compared to the alternative approaches.

Number 3 will shift resources from current consumption and private investment to a set of particular infrastructure projects. The impact on future production and welfare depends on on the productivity of the investments choosen.


If the goal is to stimulate the economy, any of these approaches should be combined with an increase in the quantity of money–that is, they should be financed by money creation.

D. Watson January 12, 2009 at 3:56 pm

One incorrect item in some people's analysis here is the assumption that every person who receives $2500 will later be taxed $2500. Most of this will eventually be paid for by the richest citizens. The poorest receive $2500 and will not be taxed later (or at least certainly not to the tune of $2500), the middle group will receive $2500 and pay back, say, $500, while the upper tax brackets will pay the balance. Thus only the rich will have an incentive to save it all a la Ricardian equivalence. If you believe in the more strongly worded versions of the Laffer curve, this would have a total negative on "the economy" unless you think the spending multipliers of the poor are sufficiently large. (I confess I haven't met anyone who believes both.)

I confess I haven't read EVERY comment, so I don't know if someone eventually pointed this out. If I was beaten to the punch, I apologize.

Previous post:

Next post: