Bastiat and Stimulus
Matt Yglesias has a smart post up on Frederic Bastiat’s “What is Seen and What is Not Seen” essay, noting that ” the correct way to understand it is as precisely laying down the theoretical conditions in which stimulative policies do work.”
He goes on:
The bulk of his early cases proceed by referring to a number of situations in which apparently wealth-creating activities are not in fact creating wealth because the money to finance them has to be taken away from somebody else. This, of course, is precisely why fiscal stimulus doesn’t consist of simultaneously increasing taxes and spending. Instead the idea of stimulative policies is to either borrow money or else actually increase the stock of money. This basically covers the points about broken windows, demobilized soldiers, and public works. And note that on public works, even Bastiat explicitly says that increasing public works spending during hard times is smart policy even if it merely shifts prosperity from the future to the present.
I agree with all of this. The entire point of stimulus spending and expanded monetary policy, as opposed to simply paying people to break and fix windows, is to use borrowed money and an increase in the supply of money to boost consumer spending, financial lending, and eventually hiring. I think Bastiat makes some important points about demobilization and I think here and in other works he makes a strong case against protectionism, but I see nothing in this essay or elsewhere that really lays the groundwork for a broad, anti-Keynesian approach to managing a recession.
You don’t need to protect the candle-makers from the light of the sun in order to increase the money supply or spend money on new rail lines or patching up crumbling infrastructure. In fact, you don’t even need to break windows when enough of them are already broken.
Another often glossed over fact is that when interest rates are pushed to near 0, conventional thinking about spending/saving doesn’t hold
If investment is timid, taxing an spending doesn’t have to be zero-sum. Especially when everyone is rushing to buy T-bills.Report
It’s not so much that conventional thinking about spending / saving doesn’t hold, it’s more that it’s much easier to get a positive Net Present Value from an investment project when interest rates are near zero.Report
In fact, you don’t even need to break windows when enough of them are already broken.
I like the idea of the whole “shovel ready” thing. Heck, I like the idea of “we’ve got all of these potholes that we’ve never been able to fill and now we’ve finally got the funding… now we just have to put together a team to examine the bids from contractors…” thing.
My problem is that we do something like give a state a million dollars to do fix some broken windows and the windows they decide to fix are the windows of the state pension fund. Let’s give a second million. Still into the pension fund. Another million. Into the fund.
So on and so forth until we give our sixteenth million… and that finally has a percentage go to fixing potholes. So we spend sixteen million to fix a half million dollars’ worth of infrastructure (but, at least, the pension fund is full again).
It’s not the half million dollars’ worth of work that I am opposed to.Report
Not a panacea, but in the category of steps the US needs to take, like yesterday, a National Infrastructure Reinvestment Bank.Report
What we need to do is find responsible members of the business community to get the potholes fixed, so they can use the money to buy options on fourth-level derivatives of frozen orange juice futures.Report
This is a real worry, no doubt. I wonder, though, if there’s a baby/bathwater problem inherent in thinking about it this way. Or alternatively, I guess, I wonder if you think there’s a <em>tolerable </em>level of corruption which would justify deficit spending, or is any corruption too much?Report
It seems that there probably is an acceptable level of corruption somewhere.
The problem is that by acknowledging such a thing, it’s more likely to balloon than if we hypocritically say “no corruption is acceptable!” and just turn a blind eye to it if it’s under this acceptable level.Report
For me, I think the goal isn’t “A World With No Corruption.” A better goal, and one somewhat more realistic is the goal “A World Where Corrupt/Not Corrupt Is More Important Than Party Affiliation.”Report
That would be nice.
Also: a world where people weren’t so panicked about corruption that they spend 15% of the outlay trying to prevent 8% waste, and only eliminate 4% of the waste.
Really, we need better accountants and risk managers in politics.Report
This is a pretty accurate assessment of where we tend to be when it comes to infrastructure project spending.Report
A world like that would rapidly reduce to a world with no corruption – politicians won’t do it if it interferes with their electability.Report
I’m not so sure. I might end retail corruption. But wholesale corruption seems to come from a different source than retail politics.Report
What do you mean by wholesale vs. retail corruption?Report
Retail corruption would be violating a policy or receiving unwarranted benefit by the political machinery in place – I’m thinking no-bid contracts to Haliburton, federal pork for local business interests, favoritism in government contract provision based on cronyism. Wholesale corruption would be entrenching otherwise corrupt practices in the institutions themselves. Like Citizens United, the lobbying/legislature carousel, placing corporate executives/owners from certain sectors as regulators for that very sector, etc. The wholesale stuff isn’t party specific, it’s class specific (it seems to me).Report
Well what does Wal-Mart do? Does Wal-Mart refrain from opening any stores unless and until it believes it has entirely eliminated any risk of shoplifting by customers or theft by employees? No it tries to put good controls in place to limit both these things. If the level starts to increase, though, at some point they may say something like there’s no point having a store here because the benefit of the additional sales will get swamped by the cost of the stuff that gets stolen.
At some point corruption in some projects such as getting a bridge built, road resurfaced, or NASA sending a man to Mars may raise the cost so much that the benefit starts to get swamped. But that in itself has nothing to do with stimulus. If we can’t build a bridge to cross the Hudson river because corrupt construction companies have such a lock on the process that its not worth it to build….well that’s a problem but stimulus doesn’t create that problem. It would be just as much a problem in a full employment economy….moreso in fact because the full employment economy lacks the ‘free lunch’ aspect of the depressed economy.Report
“I like the idea of the whole “shovel ready” thing.”
Yes, it’s a great soundbite. It’s the kind of thing that a career bureaucrat who’d never had to have the government as a customer instead of a workplace would think was easy.
I mean, why would he find working with the bureaucracy so hard? He’s done it every day for his entire professional career. He doesn’t think any more of it than fish think about the river they swim in.Report
“I like the idea of the whole “shovel ready” thing.”
There’s no shortage: it describes everything Bill Kristol says.Report
The ‘shovel ready’ thing is a red herring. Very little of the stimulus was targetted spending on ‘shovel ready’ projects to begin with. First off, most of the stimulus happens/happened outside of the actual stimulus bill. these are the ‘automatic stabilizers’ you learn about in Intro to Macro Economics. These are things like Medicaid/Medicare spending increasing when the economy goes sour because laid off people start collecting gov’t health benefits. Unemployment insurance. Welfare, food stamps and even income taxes itself (your income tax spending automatically drops as your income does). Second even in the stimulus bill, nearly half of it was some types of tax cuts directly to individuals and more than half direct payments to individuals such as extended unemployment, help with COBRA etc. There’s no ‘shovel ready’ issue there unless you want us to believe that individuals can’t figure out what to spend their unemployment check on. Even on the spending side, a lot of the stimulus was helping state gov’t asorb the increase in Medicaid which again is not a ‘shovel ready’ issue.
While road work and such is more visible and is kind of the classic textbook illustration of ‘stimulus spending’, it was actually a very tiny part of it.Report
Your reply would be more convincing if the President himself weren’t making infrastructure spending the cornerstone of his speeches about economic recovery. It’s not like “shovel-ready” was something that the Tea Party made up!Report
No its not made up at all, but I’m just going with the hard math. Look it up. Most of the stimulus package was tax cuts, income support (food stamps, unemployment extensions etc.) and aid to the states for Medicaid…not building bridges and roads. Yes that was highly visible since you see road work every day and its easy to put a big orange sign on it in ways you don’t see signs on your local hospital, doctor’s parks, or people shopping at a store….but the math is the math.Report
I’m glad that you’re doing the math, but the fact that most of the money went elsewhere doesn’t change the fact that the public face of the stimulus was infrastructure spending. It was sold as “let’s build bridges”, not “let’s continue doing all those things were were already doing that the middle class doesn’t directly benefit from”.Report
Actually, it was sold as closer to, “let’s make sure states go bankrupt and build stuff. Oh, and tax cuts because otherwise Ben Nelson and Olympia Snowe won’t vote for it.”Report
“don’t go bankrupt” obviously. Doh.Report
OK so we are going to talk about stimulus without really talking about real life stimulus. And we are going to accomplish what exactly by doing that?Report
yes, Sir, but it was the republicans that insisted we NOT do the logical things. I remember who put bloody price supports on HOUSING. I do. And I remember who had to fucking armtwist folks to cut that in half.
Can I say, sir, that no matter whether you’re a Keynesian or an Austrian, putting bloody price supports on a market that needs to crash is A Bad Idea.Report
What price supports on housing?Report
republicans wanted to give a free handout, out of my tax money, to anyone buying a house. $16,000 apiece. That artificially inflates prices, and means that the market can’t adjust to the idea that “You was the Greater Fool. Deal With It”Report
Did they propose to increase your taxes $16,000? If not it isn’t exactly handing anyone ‘your tax money’ is it?Report
If I was to assume government would spend money wisely on infrastructure, it makes sense to build up an inventory of shovel ready projects, waiting for the next recession.
Jaybird is right though. Much of it would just go to rent seekers. California’s 100 billion rail to nowhere. Pension payoffs. Union votes. Crony capitalists.
Friends of Bush vs friends of Obama. There is a better way, and it involves keeping our money out of the rent seekers hands.Report
Friends of Bush vs friends of Obama.
Maybe. But the friends they *share* always seems to be my bigger worry.Report
Amen brotherReport
It’s not even so much “rent seekers” as it is “what we are legally permitted to spend money on right now”. The government’s own rules prevent it from just saying “hey Joe Contractor, rebuild this bridge, here is all of the money”. However, they can always make more contributions to the pension fund.Report
Any organization with a budget, private or public, almost always ranks its spending in terms of ‘risks and opportunities’ which means that it is able to quickly add or delete projects as needed if there’s a sudden change in money during the year. The stimulus bill had no serious problem with finding ‘shovel ready’ projects because only a small fraction of it went to infrastructure. The bulk of it went to individual’s pocketbooks in the form of things like tax cuts, unemployment extensions and so on.
If you told me tomorrow we were going to triple infrastructure spending (or military spending or NASA spending or science spending) I’d say its a legitimate concern that the organization may not be able to quickly find, bid out and award contracts for that many projects that quickly and as a result the money would either go unspent or waste and corruption would increase as organizations feel pressure to ‘just get the budget all spent’.Report
As he does quite frequently, Yglesias assumes that borrowing money is an effectively costless act. It’s true that the cost goes up and the cost goes down, making taking one more debt a better or worse decision, depending on the circumstances, but there really is no such thing as a free lunch. The appearance of a free lunch comes from the fact that the dollar retains its status as reserve currency of choice — however, there’s no reason to believe that it is impossible to erode that status, and we increase the likelihood of doing do with every dollar we borrow.
Bastiat’s point still holds — it would be better for us to have less debt than more, ceteris paribus. You might think taking on more debt is worth the costs under cirucmstances like this (and you might be right) — but there’s still a tradeoff, and the costs involved with the tradeoff are relatively “unseen” in precisely the way that Bastiat originally conceived (that is, the costs to accumulating debt are diffuse and difficult to assign responsibility for, where the benefits are concentrated and easy to claim credit for.)Report
+1Report
To say you’re misreading Matt would imply you’ve read the piece, but you’ve either not or ignored half of it.
Matt is well aware of the costs, he’s making an explicit point that the price is less than the value. He says directly in the post that to do so, we would be deliberately accelerating future prosperity to the present, but given the massive idling of people that want and need to work – it’s a price we ought to pay. You need to grapple with the equation, not pretend that there isn’t one.
I happen to agree with Matt. Idle people can’t be decommissioned and stored until we need them. They keep needing to eat and live somewhere and they keep getting older. Our current employment situation is a massive waste.
On the currency issue, I think you have it exactly backwards. The demand for dollars is way too high (hence people buy T-bills at negative real interest rates), normally when demand skyrockets you need to raise your prices until things settle, we should be doing just that by fulfilling that demand, if we invest the borrowed money in useful things, then we can increase future prosperity to a greater degree than what we’ve borrowed.
Report
You would think that such stimulative measures would be perfected by now, automatically applied at the first sign of recession in order to keep balance, a high level of employment and steady growth.Report
Clearly it worked so well for Japan.Report
I did read the piece, but thanks for the assumption of good faith. My point was the piece is of a kind with other statements by Yglesias, to the effect that government action is effectively costless under the right circumstances. Yglesias is portraying Bastiat’s argument in an oddly pedantic way — that Bastiat’s claims about the inefficacy of stimulus are merely about the maximal efficient employment of capital stocks. The reason we read Bastiat to this day is because of the broader point he’s making about the costs to government action, not because of a technical point about the productive allocation of capital. Yglesias wants to interpret Bastiat in this narrower light, precisely to get around to the idea that there is a (approximate, effective, almost) free lunch available due to unemployed resources, and that we just need the will to overcome the idiots who don’t see this.
To be fair, I don’t think Yglesias actually believes that there is such a thing as a free lunch. But, being a frequent reader of his blog, I am often troubled by the way that the costs to government action consistently melt away from his considerations of various public policy proposals. I think this is a deliberate choice of rhetorical tactic by Yglesias, and I think he uses it to try to delegitimize the opinions of those who bring up the “hidden costs” of government action.Report
Well there is a ‘free lunch’ in the sense that if you have a lot of people and resources sitting around doing nothing, you can ‘do something’ (like make lunch) without having to add any other people or capital.
The classical assumption is that the economy is always at full employment. If you hire some people and some machines to build a bridge, those people would have otherwise been building something else so the cost ends up being whatever that other project you pulled them off would have been. The <I>transmission mechanism</I> for that is prices in the economy. Wages for construction workers gets bidded up so maybe other firms scale back some of their projects. Interest rates for investment goods go up so maybe companies take the cash from those forgone construction projects, put it in savings accounts to banks which lend the money to construction firms that buy cranes and othe rmachines on credit to handle the big gov’t contract. So the question is has the bridge that you just built provided more value to the economy than all those other projects which got cancelled or scalled back?
But again this assumes recessions are impossible. They aren’t and when you have one you have unemployed resources. You have construction workers sitting around unemployed, you have businesses that took out loans to buy cranes and other equipment that sits in the yard collecting rust. Then the bridge is, in fact, a ‘free lunch’ because the only thing you really, at the end of the day, spend to build it is giving up the ‘free time’ that those unemployed resources had.
The classical orientated economists respond by arguing because reality doesn’t seem to fit their theory, something must be wrong with reality. Hence you get really strange counter-arguments such as what Paul Krugman derided as the ‘Great Vacation’ theory…..that unemployment has suddenly spiked up to 9% because workers, for some sudden reason, just decided in 2007 to refuse jobs for a given level of income because they suddenly decided its more fun to spend time at home (usually broke or near broke) rather than collecting a paycheck.Report
“Well there is a ‘free lunch’ in the sense that if you have a lot of people and resources sitting around doing nothing, you can ‘do something’ (like make lunch) without having to add any other people or capital.”
This may be possible (although, even a PSA merely encouraging those unemployed workers to find the unemployed resources – or vice versa – would cost money,) but that isn’t what is being proposed. What’s being proposed is to attempt to put these presently unused resources in circulation by either borrowing or printing money. Borrowing or printing money each have costs associated with them. Again, you can argue that the cost associated with them are not sufficient to deter us from putting those resources to work. That’s a discussion I’m willing to have, and it differs greatly from the discussion Yglesias wants to have, which begins with “Let’s all laugh at the cranks and yokels who don’t realize that the government can print as much money as it likes.” And, honestly, I’m paraphrasing that last bit, but not by much.Report
<I>Borrowing or printing money each have costs associated with them.</I>
Actually no they don’t, not here. The only thing that has a cost, a real cost, is wasting time. The bridge workers sit around not doing anything, the cranes sit rusting in their yards not doing anything. The bridge, which could provide benefit for years and decades to come, does not come into existence. The money aspect of the whole affair is just the accounting that the system uses to transmitt these changes and make them intelligeable.
So let’s break this in two:
Borrowing – There is no net cost to society. Yes the gov’t takes on a liability and in the future pays off the bonds it issues with interest. But in the immediate period workers and capital owners get income equal to the bond that is issued and savers get the asset of the bond. The total cost is literally zero if the interest rate is zero or at worst whatever the interest cost is. If the economy is in a true depression that’s likely zero or even negative but if its less dramatic than that then all you need to care is that whatever project you’re having done has at least a tiny benefit to offset the interest cost you incur.
Printing money – Same deal here. The ‘cost’ might be inflation but at the end of the day inflation can only be ‘too much money chasing to few goods’. IF you have unemployed resources then you can add goods, there will be no inflation hence no cost.
In essence, stimulus is just a response to the signals that the market is sending. The market is saying “print more money” and “increase spending by borrowing”! What other way would you read deflation and zero interest rates (or negative real ones)? So the question isn’t what is the cost of responding to the market but what is the cost of NOT responding to the market, the cost of ignoring the market’s signals.Report
OK, let’s say that I agree with the idea that a society that borrows/prints more money is no worse off on net than one that borrows/prints less (I actually don’t agree with this, but I think that’s likely to be a dead-end here.)
Still, the quantity of “net costs” does not subsume any and all questions of costs. Both borrowing and printing money have distributive consequences, in that they benefit some people at the expense of others. Specifically, some people hold dollars, and making those dollars worth less can harm them more than the increased economic activity will help them. Again, we can argue that adopting a policy with this or that redistributive footprint is good or bad, relative to the status quo, but someone still pays the costs.
What Yglesias wants to argue, and what I don’t buy, is that borrowing/printing more is beneficial to everyone. I understand that this could potentially be the case, in theory, but in order for it to be true, I would have to believe that there were an awful-lot of pretty good economic activities being foregone (and not all economic activities qualify — the unemployed are obviously better off by being put back to work, even if that just means digging ditches and filling them up again, but it’s not clear that the stimulus that results from creating that job will render the resulting inflation as a “good trade” for someone else.)Report
Charles,
no business in its right mind doesn’t use leverage. One can bitch to high hell about how much leverage America’s gov’t has, but it’s still a TON less than Goldmann Sachs carries. I do NOT say this about Iceland, who has about double or so the leverage of Goldmann Sachs.Report
This makes sense only if you buy into the classical assumption that the economy *must* always be at full employment or if deviations from full employment serve some higher good (for example, to put some Wall Street quants on the unemployment line for a few years until they find their math Phds are better put to use modelling fusion reactions or analyzing cancer clusters). This assumption, becomes pretty implausible to maintain in the face of an economy wide recession or depression like the one we are in now. It might be more plausible to apply to the relatively shallow recessions one sees after investment bubble pops like the dot-com crash of the early 2000’s.
Distribution issues I can accept. I don’t accept that you can claim benefits are coming at ‘anyone’s expense’ unless you can demonstrate the economy is at full employment (in which case stimulus can be seen to increase inflation and/or real interest rates) or you can point to some plausible method to ‘transmit’ the cost of the stimulus on to anyone in the economy. But yes stimulus probably can’t be designed to benefit everyone equally. Naturally the guy who owns a sub shop right next to a big infrastructure project is going to benefit more than the woman running a dance school. But then recessions don’t distribute their hardships evenly throughout the economy either. In this one there are many people who’ve barely felt anything while others have lost nearly everything. Unless you want to put forth some form of idealized socialism, we should do what we can to improve people’s lots…esp. if the cost is little or nothing.Report
“This makes sense only if you buy into the classical assumption that the economy *must* always be at full employment or if deviations from full employment serve some higher good.”
I don’t think you have to believe that the economy is at full employment to believe that stimulus will be inflationary. You only have to believe that the amount of positive economic activity created by the stimulus will be less than the amount of money added into the economy. Then, there’s more money (proportionally) chasing less goods.
Fundamentally, what I object to is this idea that money is neutral in a political sense, and that managing the money supply is a technocratic problem (in fact, I don’t believe that there are many relevant political issues that are technocratic at all.) Someone’s gain always comes out of someone else’s hide — fleshing out “who”, “whom” and “how much” is the essence of politics, and there’s no way around these arguments.Report
We are getting sloppy with our terms here. The economy can be summed up in the equation:
MV = PQ
Basically money times its velocity must equal price times the quantity of things produced. PQ is the ‘real’ economy meaning the actual goods and services produced by the economy, MV is the nominal ecnomy measured in money. “More money chasing fewer goods” can only mean that M has gone up but V and Q cannot go up. That means P must increase which is inflation.
But if the economy is not at full employment then by definition Q is free to increase. That would be, in essence, a ‘free lunch’ in the sense that more money gets printed but prices don’t rise and unemployment falls…happy times. On the fiscal side stimulus is roughly equal to keeping M steady but increasing V. How is that done? Gov’t is borrowing money that otherwise isn’t getting spent (low V) and spending it directly (roads) or putting it in the hands of people who will spend it quickly (unemployment benefits, lower payroll taxes, aid for states, etc.) Again the fact is if Q is not at its max., then it can increase which means you’re moving towards a full employment economy.
The question of ‘at whose expense’ misses the point. The expense is born not by becoming fully employed but by NOT being fully employed. The rest is mere accounting.Report
Just axin’: What did BHO do with the first trillion we gave him to spread around?Report
Boonton,
You make many good points. If we are going to pay for unemployment, we might as well have them do something productive.
Some caveats:
1) Complex projects like bridges, bullet trains, solar power etc are not tasks that are easily handled by any random sample of the unemployed. Indeed, they will be handled primarily by re-allocation of current professionals and skilled laborers. Obviously the market forces will eventually work themselves down to the currently unemployed, but not necessarily faster than the market forces would have absent said interference.
2) You can’t just assume the approved project is worthwhile. Take a look at the $100 billion bullet train to Bakersfield (or wherever) that bankrupt Calif is pursuing. It is quite possible politics will lead to a huge misallocation of resources — resources that caveat 1 already shows are taken from somewhere else.
3) Maybe it would just be better to have them pick up garbage on the side of the road or do clerical support for teachers.Report
The nice thing about stimulus is that it is NOT economic management. Sure if you have some worthy projects sitting on the table, go ahead and do them. For example, take road resurfacing. Unless you have a road that’s going to be put out of commission, you’re going to have to resurface it so go ahead and do it now if you’re below full employment.
But you don’t need to figure this out, if you’re below full employment boost up demand. That’s it. The stimulus package was nice in that it was broad based demand stimulation. If you were working you got some tax cuts, unemployed got extensions and the poor got food stamps and finally the states got a time out on their budget crunches. What that demand actually demands, though, is up to the market. If you’re recession is really bad, there’s no cost to any demand boost so even digging holes and filling them up is better than nothing.Report
Monetary policy’s “failed” as of this point. If you’re against fiscal stimulus, it’s incumbent upon you to say what should go there instead (even if it’s “leave it alone and let it fix itself!”)Report
Still, the quantity of “net costs” does not subsume any and all questions of costs. Both borrowing and printing money have distributive consequences, in that they benefit some people at the expense of others…
I can agree that there’s distributive consequences in that some may benefit more than others, but I disagree you can say its at anyone’s ‘expense’ unless you can identify a clear mechanism that imposes that expense. There is no inflation being caused by ‘money printing’, for example, in the last 3 years. Hence there’s no expense…Report