What is the opportunity cost of additional government borrowing?

No, it is not zero, not even if government borrowing rates were literally at zero.  Yet I’ve seen that claim a few dozen times in the last year or two, so let’s walk through some arguments that were fully standard by the 1970s.

Opportunity cost is ultimately defined in real resource terms, converted into value.  So if the government borrows more money and mobilizes robots to do some work, that means fewer robots to do work elsewhere.

So what is the marginal private rate of return on capital?  That’s a bottomless pit sort of question, but it’s not unusual to find sources which suggest a number for the average return in the range of 15% or so, see p..53 from this Stern School study here (pdf), with the median estimate running at about 12% (p.54).  The papers from the 1980s found about the same, sometimes higher.

That’s way too high, says this stagnationist!  Let’s instead cut it down to historic U.S. equity returns and say seven percent for the return at the margin.

Now, even today there are some unemployed resources.  But most government fiscal policy works through well-known, fairly large contractors that at the margin have already well-established networks of capital and labor.  So circa 2016, I don’t think that is a significant factor.  Even in more down times, fiscal policy doesn’t always target unemployed resources so well.

That means a government project faces a seven percent hurdle rate.  You may wish to up that for risk (the value of government output covaries positively with national income), and more yet for irreversibility.  Let’s say that brings us up to a ten percent hurdle rate, and that’s being quite conservative.  Sometimes irreversibility premia can multiply hurdle rates by 2x or 3x.

So that’s a (hypothetical) hurdle rate of ten percent, not zero percent.  Of course it’s not unusual for private companies to use hurdle rates of twenty or more for their investment decisions.

There are further complications if the borrowing is financed by foreign finance capital.  But there is still likely a real resource displacement in the home market as robots are shifted from one line of work to another.  In addition, foreign ownership of the debt is about one-third of the total, noting the average and marginal here may diverge.  Still, the domestic capital case seems to be the dominant effect.

You really can bicker about the right number here, and I’ve elided the question of whether some of the crowding out might come from consumption through a positive elasticity of robot supply; check the early papers of Martin Feldstein on related questions.  But if someone tells you zero percent is the correct hurdle rate for government infrastructure investment, they are wrong.

Opportunity cost remains an underrated idea in economics.


"Opportunity cost remains an underrated idea in economics."
Not just in economics.

Did not mean to demean the original post, which I think is excellent.

"Opportunity cost" is one of the great ideas of all time. It's a good illustration of James "Flynn Effect" Flynn's insight that we get smarter as our culture comes up with handy phrases encoding complex ideas.

I realized I should look up who came up with the phrase. From Wikipedia:

"The term was coined in 1914 by Austrian economist Friedrich von Wieser in his book Theorie der gesellschaftlichen Wirtschaft.[4] The idea had been anticipated by previous writers including Benjamin Franklin and Frédéric Bastiat. Franklin coined the phrase "Time is Money", and spelt out the associated opportunity cost reasoning in his “Advice to a Young Tradesman” (1746): “Remember that Time is Money. He that can earn Ten Shillings a Day by his Labour, and goes abroad, or sits idle one half of that Day, tho’ he spends but Sixpence during his Diversion or Idleness, ought not to reckon That the only Expence; he has really spent or rather thrown away Five Shillings besides.”

"Bastiat's 1848 essay "What Is Seen and What Is Not Seen" used opportunity cost reasoning in his critique of the broken window fallacy, and of what he saw as spurious arguments for public expenditure."


Better to have workers homeless and hungry and gdp lower than to build a paved road to replace the gravel road at a cost of higher taxes and higher consumer spending by the workers paving the road?

In a number of places, communities are so short of tax revenue they are fixing paved roads with potholes by using road grinders and tampers to convert the roads into gravel roads. (Normally this is the best way to prepare the road for repaying with a two-three inch base and one inch topcoat.) Converting from paved to gravel keeps gdp down which seems to be the voter mandate.

Shhhhhhh ... keep quiet about the gravel road thingie or the city of Los Angeles will do that to keep speeding Porsches from plummeting off Mulholland Drive. So far they've just been letting potholes accumulate on the winding highway on top of the Hollywood Hills for most of the century in the name of cutting down on street racing. But this pulverizing idea of yours sounds like just the thing they'd do next.

20% of LA's budget is now pension expense, double from a decade ago. Infrastructure has been crowded out.

I agree on opportunity cost being up there with comparative advantage and EMH as being true, non-obvious and non-trivial. And it's amazing how many people still argue against these basic ideas.

One of the nice things about being a financial economist as opposed to some other kind of economist is that opportunity cost is drummed into your head from the first instant.

Economics in three aphorisms:
People respond to incentives
All decisions involve tradeoffs
Opportunity costs are real costs

I normally give economists a lot of grief, but, I must say, these are just tremendously valuable ideas.

If Intro to Economics did nothing but instill those three ideas into it's students it would be an invaluable service.

'I agree on opportunity cost being up there with comparative advantage and EMH as being true, non-obvious and non-trivial.'

Almost as if the expression concerning having cake and eating it too was not a common expression in English since the mid-16th century - 'An early recording of the phrase is in a letter on 14 March 1538 from Thomas, Duke of Norfolk, to Thomas Cromwell, as "a man can not have his cake and eat his cake".

The phrase occurs with the clauses reversed in John Heywood's "A dialogue Conteinyng the Nomber in Effect of All the Prouerbes in the Englishe Tongue" from 1546, as "wolde you bothe eate your cake, and have your cake?". In John Davies' "Scourge of Folly" of 1611, the same order is used, as "A man cannot eat his cake and haue it stil." In Jonathan Swift's 1738 farce "Polite Conversation", the character Lady Answerall says "she cannot eat her cake and have her cake."

The order was reversed again in a posthumous adaptation of "Polite Conversation" in 1749, "Tittle Tattle; or, Taste A-la-Mode", as "And she cannot have her Cake and eat her Cake." From 1812 (R. C. Knopf's "Document Transcriptions of War of 1812" (1959) VI. 204) is a modern-sounding recording as "We cannot have our cake and eat it too."' https://en.wikipedia.org/wiki/You_can't_have_your_cake_and_eat_it

Troll. But see Shiller on excess volatility as a starter.

EMH is not true unless you assume a bunch of things that are never true. But its predictions are a useful point of reference. I agree about the other two.

I will try to limit myself to 2 anti-EMH comments, this being the second.

The EMH is false because it was disproved. Remember the idea that real science was supposed to be falsifiable? The EMH was falsified. It's easy to google anomalies and exceptions that show it so.

Fama's position now is that "it's only a model" so it's good even if it was disproved. If that floats your boat, I guess. You can again rank it as a basic truth.

I've had season tickets to the Chicago Blackhawks for a decade now. Opportunity cost is the asshole in the back of my head who tells me what I'm REALLY paying for these tickets.

And I'm a Cubs season ticket holder too! O opportunity cost- keep stinging with your lash!

I mean, did you see that game last night? Coughlin asks for time on a 3-2 pitch, doesn't get it, he jumps back in and strokes a single, two ribbies, ties the game. Then 37 year-old catcher David Ross lays down a two-out bunt to push the lead run across.

In the 7th inning, the Cardinals sent five men to the plate. Single, K, Homer, Single Double. 4 hits in five batters. And the inning was over. 1 run. Cuz David Ross picked a guy off first, then a great relay from Coughlin to Russell to David Ross allowed them to cut down noted douchebag Matt Carpetner with a humiliating tag/facewash at home plate.

In the 12th, phenom Kris Bryant whiffs when a flyball woulda won it, then Rizzo wins on a walk off walk. Hilarious. Can't make this shit up. Suck it Cardinals.

Oh yeah. And opportunity cost.

Oh, and, great post Tyler. You should write about economics more. Seems like you're getting bored with it.

Regardless I think the US is at the border of what is feasible when it comes to goverment borrowing, it's interesting to see the idea of the economy as a zero sum game alive and kicking.

On a tangential issue, as I read the line "In addition, foreign ownership of the debt is about one-third of the total" I thought, when Trump states that as President he might not honor our debt obligations, he's portrayed as a mad man.

When Paul Krugman writes a column title "Debt Is Money We Owe To Ourselves" implying the same, well...crickets.

Paper cups, sticks, bags. Americans complain even about free money.

"Americans complain even about free money."


Effectively free. I would have to pay a hundred percent interest rate if I were to borrow money and people complain they have to pay zero real interest on their Brobdingnagian deficit. My blood boils.

"Now, even today there are some unemployed resources."
This is a huge understatement, given the obvious trends of excess capacity seen over the last 50 years.

"Compounding the problem of excess capacity and stagnation is the structural maladjustment that results from neo-Keynesian remedies for relieving excess capacity, which call for stimulating the economy through deficit-financed government investment. Government spending, especially within the growing movement toward privatization, seeks to exhaust the economic surplus fund by closing the gap between productive capacity and now increasing public-sector consumption. As early as 1957, Paul Baran pointed out that such stimulation within the monopoly capitalist system encourages business overestimation of demand elasticity, which in turn encourages production beyond the limits of consumption, thereby amplifying the economy’s tendency toward stagnation. The result is simply a circular drive for further government stimulus of the economy. Moreover, financing unstructured investment through government deficits, especially in non-capital-stock-increasing investment—such as military armaments and technological research—would result in a precarious inflationary overhang." -Mary V. Wrenn, "Surplus Absorption and Waste in Neoliberal Monopoly Capitalism"

For those of us not already in the know, could you describe the excess capacity in question?

What is the opportunity cost of NOT increasing government borrowing?

Libertarians always want to see extreme restraint of the state apparatus: having lost sight of the fact that this large state apparatus is a necessary condition of the modern monopoly-capitalist social order. The capitalism that is with us today did not spring fully formed but has evolved over the centuries. Institutions are developed, such as the Federal Reserve--functioning as the lender of last resort--in response to real contradictions, which serve to maintain the system as a whole. Asking about the opportunity cost of government borrowing is non-sensical in that, if no borrowing were to take place, capital overhang/lack of aggregate demand, and strife, would be the order of the day. "We are all Keynesians now."

"What is the opportunity cost of NOT increasing government borrowing?"

I'm sure one of these days, someone will test it, but till then it remains a theoretical concept.

A closed school. A closed hospital. A closed road. A poisoned lake.

Surely, we need not propose 100% allocation of GDP to government or literally infinite borrowing to understand that government cannot do all. But sometimes the opportunity cost of NOT increasing government spending can be rather high.

But I guess that might depend on whether you believe that private markets will actually build all roads, bridges, schools, etc., which have the expected ability to increase economic well being and potential of the country.

So long as population is growth, deficits are not a problem so long as the debt-to-GDP ratio is roughly stable, accounting for reasonable business cycle fluctuations.

Recall, wanting small government is not the same as wanting no government.

A closed school means that there are fewer kids and the district is rationalizing costs. A closed hospital is the result of out of control health spending or mismanagement. A closed road is due to lack of economic activity or residences in a place that used to have it. A poisoned lake needs to be fixed.

Deficits allow governments to spend more than they are willing to tax. Essentially deficits are the overhead of corruption. It is amazing how well governments can run when they are forced to live within what they can collect in revenues. Regulations, expenditures and operations end up reflecting what needs to be done, not some pipe dream of a politician making a name for themselves.

If you can't collect the taxes, don't spend it. And yes, there is a limit which you can tax, which is a natural limit of the size of government.

Those would be good reasons to close a school or a hospital. I'm referring to the case of it happening when such situations as you refer to do not apply.

Question: If you could buy and hold on two share prices, each firm starting from an identical place, but one which committed to never take any loans, borrowing, etc., and one which explicitly committed to useful of diverse financial instruments to optimize the profitability and continuity of its operations, which share would you buy?

Why should it be any different for countries? It's just, you're missing the market discipline signal, so better management mechanisms may be strictly required ... but I say this as one D Drumpf and HR Clinton square off for top post in the land, so what hope is there that voter/civic engagement would be that much needed endless prodding which a firm might expect from the accounting department. God knows, much media seems quite OK with blank checks so long as it goes to the party and priorities they support.

If Republicans actually buy half the ideological propaganda they've been fed for decades (not saying it's bad, but it's ideological, it is propaganda, and it has been decades), then it seems at this juncture, it wouldn't be horribly surprising to see half the party jump ship and go Libertarian, in the case that trying to take back their own party doesn't work very well. We'll see ...

Total Opportunity Cost vs Net Opportunity Cost.

Too many people confuse about the difference between these two types of opportunity costs. Especially Economists.

Tyler linked to this confusion in a post from a few years back, stemming from a Robert Frank article. The follow up was a discussion of the many interpretations of the term. Tyler never came down to favor any of them.

The idea of opportunity cost, as stated above, is very powerful. It is a shame the field cannot harness this power with more clarifying language. It is our job to, especially for adoption of its more widespread use and understanding

What's worse is the word "Net" has a different meaning in economics and finance. Yet, they are suppose to be two closely related fields.

I think basically it gets taught in lesson 1 or 2 in intro, maybe a couple basic questions come up on an exam or quiz, and then it's basically never heard of again.

For what it's worth, I'm constantly trying to drill it into my product management team. You have to go beyond the cash cost of the product or initiative, and also beyond the hurdle rate. You have to think about whether you could be doing something more useful if you didn't start this project. The biggest challenge is that people are very, very bad at hearing the dog that doesn't bark.

Additionally, who is to say rates won't go negative in the United States in the future?

They can be -2% and all of a sudden .39% seems quite high.

"They can be -2% ..."

Hmmm, I'm not convinced rates can get that low. At some point the incentive to find other places to put your money (say a chest buried in the map on a tropical island) would seem to have a lower effective cost.

Yeah, may be 1%. I'm pretty sure I could guard $1 billion worth of gold (or currency) and keep proper accounts, etc., for under $10 million a year. By the time you're getting to $20 million a year, you'll have non-currency unit-of-account competitors easily underbidding each other to store value.

Sure, theoretically various rules could invalidate such approaches. But I don't think people would stand for that. We tolerate fiat currency because it works fantastically conveniently in many ways. Put a 2% tax on it and you may as well just print a new currency, imo. Maaaaybe the USD's special status and all, it could be an exception.

Real short-term rates have been negative in the USA for the past eight years.

I am not sure this is right but I would reason as follows:

This seems to imply , the government could use the borrowed money and invest in index funds and get the historical rate of return ( 7%), but governments don't do that anyway ( except maybe Norway), they instead simply not borrow, so this can't be an opportunity cost for the govt even though it could be for a private firm.

If the opportunity cost comes from the govt. competing with the private sector for employees and machinery, then the economy is more or less at full output and employment and this govt stimulus borrowing makes no sense, it should wait until there is less than full employment and investment otherwise it will create inflation.

If the economy suffers from a lack of demand and investment and is not at full employment , then this may make sense and the opportunity cost is just the cost of borrowing.

As far as the return on investment. If the govt borrows say $ 200B to fix a bunch of potholes and bridges , OK, road and bridges are therefore improved but how is the return on that measured ?

If you are fired because too many people do not have jobs and welfare is cut to not borrow to pay people to not work, what is the benefit of you becoming homeless and hungry? (If you will not stop consuming at your current rate if no longer working, why are you taking the opportunity to earn and spend from someone younger and unemployed?)

No opportunity cost still applies here. Actually this is easy to work out, the opportunity cost is exactly what the shares return. The government could have left the money in the private sector and the private sector would have got that return. In fact probably this is value destructive due to Government agency costs (reference General Motors).

The more sophisticated argument for government borrowing to buy shares is that demand for shares is artificially depressed due to panics and so like a bank run government action is needed to stabilise the market, if this works then Governments will get above normal rates of return. But it's hard to make that argument today.

What perhaps a better stimulus idea is money printing rather than borrowing by central banks when bonds returns are so low to buy shares. By the "hot potato" effect this puts money into the economy perhaps more efficiently than bond buying, where a lot of money ends up in reserves. It should also lower the cost of capital for companies encouraging them to invest more.

@ChrisA. What does that mean leave the money in the private sector ? The government borrows and offers bonds : the 10 year note yield is 1.5%. If the private sector prefers to invest in index funds at 7%, then why do they buy the bonds or alternatively why doesn't the bond yield jump to 7% ?... Because the risk is not the same, the 7% return is not guaranteed like the bond yield is. I think opportunity cost does not factor the return risk

Yes opportunity cost does take into account risk.

The logic is sound. But if private markets refuse to do something which must be done (cannot internalize gains, investment is too long term to get investors, etc.), then something the government must allocate resources in a way that will have a non-zero crowding out effect.

I don't think we should be concerned that the crowding out effect should be zero. Rather, that costs exceed benefits (and allowing that some people have different perceptions of what should be considered in assessing costs and benefits - financiers having perhaps the most narrow definition of all).

What is the opportunity cost of $1 trillion in wages and benefits paid to American workers to replace 75+ year old water and sewer pipes, 75 year old end of life bridges and 50 year old bridges that are too narrow or lack carrying capacity, eliminate railroad grade crossings and raise freight rail mainline speeds to a minimum of 120 kph, upgrade all major ports to superpanamax, upgrade rail capacity of these ports to quickly dispatch all cargo to intermodal depots and warehouses, run fiber to every address in the US, ...

And that is $1 trillion per year investment in wages and benefits paid to American workers over and above the labor cost run rate of the past decade?

What would be the opportunity cost of not ensuring $1 trillion in added consumer spending driven by $1 trillion more in working class worker incomes?

Is the reason for not investing one trillion per year the fact the labor force is currently overcommitted in today's economy with zero slash labor?

I'm saving to put an extension my house, and so I can buy a more flash car next time. Please don't lecture me about sewage and bridges and stuff.

Ok, don't complain when the sewage backs up into your basement and you are reduced to working two part time jobs because all construction being shutdown to not hike taxes has rippled through the economy to your job, and put the boat builders out of business leaving only the guy who works for Larry Ellison.

I remember my youth in Indiana when tax hikes to fund bonds to build schools or roads or water were almost annual affairs and perhaps the reason adults voted every year, to vote no, or yes, on tax hikes, but everyone went from bitching about the latest tax hike into talking about their new boat, or new RV trailer, or new cabin on the lake or their car with big fins. I lived in a working class town surrounded by farms that grew everything I ate except for the bananas. Nixon appointing Earl Butz and then Reagan pretty much ended both.

Last time I knew anyone talking about, and buying a boat, Clinton was president. But never as many talking about, and buying, boats in the 35 years I've lived in NH as when I lived in Indiana in the 60s and 70s.

I'm always suspicious of these private rates of returns, especially in the aggregate. Show me a relatively safe private investment returning 7%, let alone 10% these days...

You are looking at retail returns not wholesale ones. An investment by say an oil company in a refinery perhaps gets a 20% return but you can't invest in the refinery, but in the oil company shares. You also have to be aware of survivorship bias when looking at private rates of return, poor investments are usually abandoned very quickly.

Exactly. This is the internal profit, before all risk adjustments, employee bonuses and taxes on profits, etc are paid out. Obviously, the net profit is substantially lower.

"But most government fiscal policy works through well-known, fairly large contractors that at the margin have already well-established networks of capital and labor."

This looks like hand-waving. None of these contractors have laid off staff in recent years or pared back hiring among groups who are presently unemployed?

Same goes for capital. Is there reason to believe all all producers of capital goods are fully employed and all businesses that lease out equipment to contractors don't have much spare equipment in their inventories right now? If the answer to both of these questions is yes, then we should expect the economy to be in great shape right now.

"You really can bicker about the right number here, and I’ve elided the question of whether some of the crowding out might come from consumption through a positive elasticity of robot supply; "

So, what will be the gdp of an economy where robots have replaced all workers if government is not borrowing an amount equal to gdp to pay welfare to workers to buy robot production to ensure gdp is not Zero?

Or will robots replace both workers and consumers?

What will Robot Walt create for cartoon characters for film and theme park rides for robots to enjoy?

What will Robot Zuckerberg create for robot social media, and how will it monetize it? Or will robots have a sexual identity so it will be a she, or a he?

Will GE robots mate only with GE robots, or will sex between GE and Hitachi robots be socially acceptable? How expensive will the weddings be?

The article confuses the opportunity cost of spending with the opportunity cost of borrowing. The cost of borrowing is near zero and below zero for many governments. There is no law that says a government has to spend what it borrows. At negative rates they would make money from borrowing and not needing to pay it all back. If you believe in price signals, negative bond markets are telling governments that's what they should do.

Just like every indicator, every signal was to buy a bigger house with more leverage in 2005.

It is the epitome of stupidity to think unprecedented low interest rates are going to remain that way. It is the definition of madness to think that this money will not have to be refinanced at a higher rate in the future. The government doesn't pay off it's debts, it simply refinances. So the costs of the borrowed money are interest payments on it to perpetuity.

Borrow for one period, do nothing with the proceeds, pay back. At negative rates, you profit.

That's completely different from speculating in housing or borrowing that only makes sense if you can refinance at a similar rate.

You are kidding right? We are talking about government borrowing money.

This is free money to follow some politician's pipe dream. It isn't free, and it will be wasted.

It is free, and it would be wasted and used well in the same proportions as any other money the government gets.

re: opportunity costs of borrowing...

The wealth government borrows must already exist, or it can't be borrowed and used. In simplistic terms in order to help you understand the point of the article, that wealth can either be loaned to the government, or it can be loaned/invested in something else.

The wealth which is loaned to the government (even at a 0% interest rage) wouldn't have alternatively been buried in the ground or destroyed, it would have been loaned elsewhere, or invested in something, etc... So the opportunity cost is whatever benefit would've accrued from that wealth if it was spent/invested privately, rather than by the government. If the project the government wants to spend it on doesn't have a better return than whatever the opportunity cost is, then it makes more sense for the government to not borrow the wealth in the first place, even at 0% interest rates.

Also, I'll just point out that negative nominal interest rates aren't the same as negative real interest rates. You can have the former without the latter.

Borrowing money does not mean anything for real resources. Money is just account balances in banks. Buying things reallocates real resources.

At the risk of stating the obvious, when money moves from I to G, rather than from I to P, then it is G who chooses what to buy, buys them and consumes real resources, rather than P. P is now unable to.

The money/accounting involved is just how we track the flow of where real resources will end up going. Sure, theoretically G and P could both have been planning to just let it sit there in cash, but in the real world the probability is too low to matter. After all, they borrowed it to spend/invest it, otherwise, why borrow it in the first place?

What if the fiscal policy causes people to shift from crappy low-paid work to higher-paid work? I can't see Donald Trump touring blue-collar country and saying, "Actually, you've just got to accept that these jobs are the best you can possibly get, folks".

The MP of workers is not constant. Some workers are ZMP w/o financial engineering but median with the conversion of NINJA loans into AAA rated ABCP. Similarly some workers are ZMP w/o fiscal spending (even if through govt contractors), but median MP with.

Your earlier arguments about the NPV and public choice/ political economy issues around government spending were better. These substitution effect type arguments are easy to discard.

In any case, if you believe that the frontier for marginal spending is not very attractive, how about tax cuts?

p.s. The NPV and hurdle rate arguments are not isomorphic, btw. The NPV argument is more like: at the -1 sigma level, the rate of return is -ve. That is much more convincing than positing a 10% hurdle rate because of historical equity returns and 1980s papers and suchlike. The 20% IRR projects? All-equity financed for SMEs that have to have a large margin of safety. Does J P Morgan set a 20% hurdle rate for its Treasury trading desk? If not, why not?

And how to account for risk? Doesn't need to be more complicated than solving for OLG dynamic efficiency. What is the NGDP trend growth expected/planned now: 4%? 4%+ epsilon is your number. Oh call it 5% for historical reasons.

Not every investment increases productivity and, thus, economic growth. Indeed, economists used to focus on "productive capital" as the engine of economic growth. Alas, that focus has gone out of favor: investment is investment in the new economic order. In Adam Smith's world, owners of capital invested in factories and machines that produced goods to be sold to customers. In today's economic order, owners of capital speculate in real estate and stocks and bonds and even in entire companies, buying and selling at a furious pace to capture the ephemeral gains of the moment. I recall a time when "speculator" was an epithet. Today, it's a compliment. Of course, the speculator doesn't care about public investment in pubic goods like infrastructure because it doesn't affect the speculator's return. Even the boy wonders in Silicon Valley have little affection for public investment in public goods: why would someone in the business of advertising. America has shifted more than the means of production to China, America has shifted the very heart of capitalism and the engine of economic growth and prosperity. Those crazy Chinese, investing billions and billions in productive capital including infrastructure. Fortunately for us, the Chinese are a quick study, and soon enough they too will mimic the ways of the owners of capital in America.

Excellent post!

I'm not sure I agree with the unemployed resources analogy.

The argument seems to be that supply markets are sticky. A gov't road contract, say, cannot increase employment because there are only a few road contracting companies and they cannot increase capacity quickly. So then that should be easily observed by spikes in such contracts. Just like with uber surge pricing, a sudden influx of demand in a supply that cannot be expanded will cause price increases and higher income for those lucky enough to be employed there.


1. How likely is the company to be fully employed in a depressed economy to begin with? For example, I recall CNN running a special on the stimulus program a few years ago. They zeroed in on signs. Some states put up big metal signs by stimulus projects letting taxpayers know this was part of the recovery act, other states did not. CNN estimated the signs cost about $1M collectively for all the projects. They ended with a graphic that said something like "Signs $1M, jobs 0". Really? Are we to believe sign makers around the entire country were running at full capacity or that there were only a handful of mega-companies supplying metal signs? The additional orders provided absolutely no overtime, even temporary hiring, or even made a marginal difference in avoiding a layoff? The gov't here does not have to be exceptionally good at targeting unemployed resources. Signmakers may not have been the resource with the most slack capacity but it was only sufficient that they had some serious slack.

2. Let's say the contractors are at capacity and the contract simply produces higher prices. The real use of unemployed resources comes from demand. Since the income earned from the contract is free to be spent however they wish, signmakers would likely find the areas with the most unemployed resources would have the lowest prices. Demand will find the most unemployed resources on its own

Not bad at all about signs. My understanding is that in Canada, the governing party had spent about $750 million advertising the "economic action plan" in their party colours, using public money. Yep. You got it. $25 per Canadian for partisan advertising in partisan colours, purporting to be the government out to help us when they were actually doing next to nothing.

Most projects? Couldn't see any evidence of any actual work, but lots of signs explaining how the government was doing its thing. Oh, and no one costed in sign removal at the end of the program. About 5 years later, there are still signs dotting the countryside here and there ...

There were lots of curling rinks for very small towns though, at a good few million dollars a pop.

On the east coast (Atlantic Canada), one town asked why another town got such benefits, but they didn't. The PM basically said "well, they voted for me". The PM's party got completely locked out of the entire Atlantic Canada region. I think it didn't help that he called them lazy asses.

This is also why I believe tax cuts have little to no stimulative effect. They don't shift any real resources in the economy. If government spending is unchanged, tax cuts just change how it is financed - the dollars returned to the private sector as tax cuts are soaked right back up when the new debt is issued.

Let's keep in mind 'government spending' to most people means accounting's definition of spending....i.e. adding up all the checks. In economic terms it means only the government purchase of goods and services. The government sending out an extra Social Security check would not be spending since the actual stuff purchased would be by the individuals who get the check.

For all the talk about 'shovel ready' stimulus projects, even the Obama stimulus was very little actual gov't spending (less than 50%, probably less than 40% even). It was mostly direct and indirect tax cuts.


What is the opportunity cost of mobilizing an inactive resource? Do macro planners account for my loss of enjoyment of long morning walks and Jerry Springer marathons?

No they are probably looking at things like the labor force participation rate plus industry centric metrics on utilized and un-utilized capacity. If your local economy started booming and your boss started offering people pay to work overtime you yourself would start to consider would coming in an hour early rather than taking a morning walk be worth it or not?

Lots of very sensible and relevant things flow from what you say.

But I was busy bemoaning how macro planners don't generally too concerned about those aspects of well-being which are not easily packaged into a formal transaction.

The best stuff in life is free, but it's not even on the radar for the people responsible to help more and more get the best stuff in life (and make the nation stronger and stuff too).

The 'infrastructure' that this borrowed money will go to is topping up the bankrupt pension plans of most levels of government. And hip replacements for the aging boomers.

What I find very interesting in the comments here is how the notion that money is not worth anything is becoming accepted wisdom. If you have an asset and cannot collect rent when someone else uses it, the asset is not worth anything. Tyler is proposing that the asset actually has worth and better rent it to someone who will generate a return and be able to pay rent for it.

This is happening as I expect. The worthlessness of capital is already evidenced in specific sectors or micro economies. Pray that it doesn't spread. This isn't some manageable inflation, but simply that capital has no value. A very strange and third worldish situation where the flows are hard and minimal, and the accumulation has no value.

Huh? Under that logic mortgages, car loans, and cash savings all face the same bar. So much waste everywhere!

And seriously, you are arguing that the efficient market hypothesis is wrong, because you believe that people are inexplicably giving money to the Gov at 0% when they could just sit back and get the 'historic U.S. equity returns' of 7%!

The correct counterfactual is that those who are currently buying bonds would allocate that capital to some other low-risk-low-return asset with a return of around 0%

Yes indeed. Borrowing money doesn't make you rich. Investing money in a return generating enterprise does. I buy trucks because spending $30k generates $300k in revenues. When I was on a payroll, I bought a car so I could show up for work.

As for the incentives to lend money to governments at low rates, maybe that indicates that other investments essentially do the same thing at higher costs and risk. I gave one of my guys a raise this year as he works through his apprenticeship training. The costs of investing in his skills is a 50% marginal tax rate on his wages. To pay for the increase requires higher sales volumes, with the attendant risks and costs. The government gets more money from my business than I do.

It could very well be that it would be cheaper for me to put my money into a very low yield government bond. I know that it is for many business who operate at far lower margins than I do.

As usual with these things, governments build these ridiculous incentives, find that it doesn't work very well, and the obvious solution is to borrow vast sums of money.

Don't forget that a lot of "investment" in government bonds isn't in the sense you think of.

Say, you want some liquidity. You want to know that, any any moment in time, you can get your hands on $50 million in case ... in case whatever. It just makes sense for your business to do that for whatever reason.

So, you sit on treasuries at 0.25% instead of cash at 0%, because it's marginally worth the time of those expensive people to roll over treasuries regularly so that at any point in time, you'll have $50 million on your hands in very short order.

Also, many will hold 0.25% certain return due to risk-management regulations.

But in neither case do I believe the investor to consider this really an "investment". They buy these assets as a part of managing risk or liquidity, not because they're supremely attracted by a 0.25% return. And, what of when it's a -0.25%? Well, if that's higher than the -0.5% being paid to cash, then they will take it.

Once all projects with IRR > 5% have been done though, one wonders where future growth or activity will come from.

Of course they do. Why wouldn't they?

Hurdle rates dont make much sense in considering debt-financed investment in public goods. A public road or railway built might never recuperate its cost via income streams generated by the investment, yet it might have a considerable positive economic value nonetheless via positive externalities. If you wanted to compare opportunity costs one would have to compare the utility of a public good (over the time of its use, including externalitites) vs.the potential utility lost from having to pay taxes in the future for the investment in the public good today...

So why not fund it out of taxation? All levels of government take somewhere above 40% of GDP already. Do you mean that after collecting those trillions of dollars they can't afford to build a road?

It does if you have just one government agency that collects all taxes and does all government spending. However you don't. A road may increase sales to businesses along it, that results in tax revenue to the IRS on income taxes, Social Security on SSI taxes, perhaps the Federal gov't collecting customs duties on stuff coming in at ports, the local government that's collecting property taxes and the state government collecting sales and income taxes. The road itself might be funded by a mix of state and Federal dollars.

So when the state 'transportation fund' is running dry and people are debating increasing the gas tax, does anyone go to Social Security or unemployment or the state's general revenue or the local gov't property tax collections and remind them that a portion of that is coming from that road so they should fork it over?

Analyzing 'government' as though there's just one government in any developed economy is as silly as pretending you can model 'business' as though it consists of a single firm owned by a Mr. Monopoly like character in a top hat.

In the U.S., the number is 33% not 40%. The federal gasoline excise tax of 18.4 cents per gallon -- which is supposed to finance federal highway construction and maintenance -- has stayed the same since 1993 and has not even been adjusted for inflation.

And this does not even include the regulatory subsidies awarded to public debt. Just take those away and its rates would be much higher.


I read this as agreement that more borrowing is possible in the US, but that care should be taken. That's pretty much my position. I don't think the now dying idea of college for everyone is realistic as an investment because both low return and opportunity costs. On the other hand the rising idea of a trades path might satisfy both. etc.

Price is price.
If a business can borrow at 1% for term, and invest that money in such a way that it exceeds that cost of capital on a risk-adjusted basis over the same term, then it should do it! We don't complain that the business will consume resources (bidding up price of labor, materials, robots, whatever) because the price signal ensures the marginal cost (incl opportunity cost!) of those resources is accounted for. Why is government any different?

I read Tyler as saying government should, within limits. Perhaps someone more extreme would say government should not invest at 10% return because by assumption they are crowding out a private business that would make 11% return ..

You are describing two worlds. If I invest money into an enterprise it better generate a return. If it doesn't, it stops. The business ends. That discipline keeps a limit on the stupidity.

When government 'invests' it is money in addition to the 40% or so it already takes out of the economy to do something it should have been able to do with tax revenues but can't because it can't control costs or act rationally. Government is a service to the economy. The whole problem driving this desire to borrow is that it can't generate enough revenues to satisfy the pipe dreams of every piece of scum out there, is already slowing the economy with what it sucks out of is. So the obvious solution is to borrow lots more money.

When government stops, when the tax man goes home, the police stay home and I don't have to deal with the list of regulators when they run out of money because of their foolishness, then I will consider your example.

There is absolutely nothing wrong with demanding good accounting and positive ROI for anything government calls "investment."

But you know, foes often don't do that either. They just face those who assume positive ROI with an assumption of negative return.

A lot of public sector activities are justified on a basis other than investment return.

But the basic logic of not culling inefficient allocations is still relevant, I think. The market does that on its own (although certainly there can be many other reasons for business failure), but in the public sector, it is not always obvious which feedbacks are would lead to closure in the case of inefficiency, unnecessary offerings, various failures of management, etc. Then there's the question of the extent to which the public actually can access influence through their elective representatives, for better or worse.

Tax base size and other preferences of citizens also affect this stuff. Say, it costs money to have flowers in all the boulevards, but it's kind of nice. And certainly not many people will just go do that on their own, but for a couple dollars a piece it's quite the deal. And livability is related to tax base stuff (talented people might generally choose to live in places that are nice to live in).

Not to mention that technically health is public sector in many places, but it appears as though there must be some pretty significant efficiency savings relative to the US system, although it is undeniable that the high prices also attract talent in many health-related specialties.

Are you suggesting public debt, to be repaid by our grandchildren, should be taken in order to have flowers in the boulevards? :-)

Well if you put it that way ...

I'm thinking more like, someone comes around asking "do you support a $2 per person average municipal tax increase to have flowers in most of the boulevards in the summer?", I'd be OK with that. Some people try to tell me that it's a matter of ignorance because I don't own property or pay property taxes, but I'm not ignorant - I know those taxes are included in the rental price.

And for concern about some people paying a large share of taxes, well, I'm sure most of them might not mind to drop $10 on flowers in the boulevards (assuming that if you pay lots of taxes, you've still got a decent bit left), while another person might put their line around $2, which is probably basically reflected in shares of property taxes.

The city I'm thinking of, plausibly it could even be argued to take on debt in relation to it, because it's recovering, in a sense, from perceptions linked to a fairly industrial history (partly ongoing) which may limit investment from nearby wealth centres (which also makes it more affordable...).

If it can attract additional investment, for example in relation to being proximate to some other centre of wealth and population, maybe it could pay off in some cases to do so in a deficit scenario with a view to the long-term tax base?

I don't mind paying the $2 for flowers either... that is as long as those flowers are competitively priced :-)

No. Just, no. The opportunity cost is priced by the interest rate you could borrow at. And that's close to zero.

Mic drop.

The spending, Thomas, the spending.

Pick that mike back up!

I assume that the fabled "good" kindergartner teacher who adds $100K lifetime earnings for each student would be a "spending" that qualifies?

Don't you mean the best alternative use of the resources used by the government?

I think your estimate of 7% marginal returns involves more than a little hand-waving, and the 10% hurdle rate involves heroic amounts. Cutting an average rate of 12% "down" to a marginal rate of 7% is not much of a cut - it may be no cut at all. Are there really no low-return projects in the private sector?

Oh, and you argue that You may wish to up that for risk (the value of government output covaries positively with national income), and more yet for irreversibility. Yet private output faces exactly the same risks.

One way to conceptualize things is to consider the following:

Here we have a robot. Both the government and ABC, Inc. have some productive use for it, and must submit bids to buy it. Because the government has a lower borrowing rate than ABC, it is able to outbid ABC, effectively earning a lower return than ABC would have. But the return is lower only in a nominal sense. In a real sense - the contribution of the actual physical product to national wellbeing - it need not be lower.

What about the actual economic resources put into action as a result of that? If the private sector won't, should the government?

So if the government borrows more money and mobilizes robots to do some work, that means fewer robots to do work elsewhere.

Well, if the supply of robots is fixed, all are in productive use, with the least useful returning 7%(!) and there is no slack capacity for making more.

I’ve elided the question of whether some of the crowding out might come from consumption through a positive elasticity of robot supply

I don't understand this statement. Perhaps TC could clarify.

Finally, suppose a private company could, somehow, borrow at 1% and invest the capital at a 3% return. Would that be wrong? Destructive? A bad idea?

Statist baby-boomers want us to extract all existent public borrowing capacity, and leave nothing of it on the table for the future


Cowen should be in the Olympics, in the gymnastics competition.

I thought Baumol settled this in 1968... https://www.jstor.org/stable/1815533. Use the private sector return.

The implication seems to be that higher opportunity costs are a good reason for governments not to spend on infrastructure. But I'm not sure that's right in present circumstances.

Increased demand for robots and inelastic supply should lead to higher prices, which is mostly bad. But presumably the primary resource we're talking about here is human labor. Increasing the price of labor is more or less just raising wages. From a macro standpoint that's a good thing in present circumstances, right? Higher wages paid for through progressive taxation (now or in the future) is a form of fiscal stimulus and it would reduce income inequality, not to mention that higher wages in general for people building infrastructure = a higher standard of living lower middle class workers.

Bracketing right-wing ideological objections, mightn't these upsides balance out the downside of the opportunity cost that you identify?

Opportunity cost remains an underrated idea in economics.


Indeed, I was taught that all costs are opportunity costs. Buy this and you can't buy that.

Isn't there another opportunity cost to consider for public infrastructure spending - if you put off spending when interest rates are low, you have to spend more when interest rates rise. The infrastructure won't fix itself in the meantime... plus opportunity costs exist in the sense of early/preventative maintenance versus waiting and having to perform more extensive repairs. Ironically, the private sector understands both of these better than than many market oriented commentators.

The post is probably too short to cover this issue adequately. That said: the post only discusses the opportunity cost of building, and not the opportunity cost of not building. Furthermore, the characterization of opportunity cost itself is too narrow: the post assumes robot scarcity (presumably a proxy for labor scarcity), but makes no offsetting allowance either for a) idle labor capacity or b) the absence of the infrastructure that needs to be built, and the cost of its absence.

Finally, why is the hurdle rate not altered by the existing interest rate environment? If the interest rate environment is exceedingly low, it almost certainly is an all-in sign of slow economic growth, low demand for capital, and likely low fertility rates. Not to mention the market expectation that the factors behind the low interest rate environment will continue.

As Larry Summers has pointed out (he's not the only one) designing for, or building, during uncertainty is alot easier when current long term rates are low, and one can have a reasonable expectation that certain, specific public infrastructure--again, which we need and which doesn't yet exist--can provide a long term positive rate of return.

As a coda: I'm no longer convinved the US government has to issue debt to build long-overdue public infrastructure. If we are going to take delivery of high tech public infrastructure that provides services, enhances GDP, promotes health and well being, the government can simply print the capital. In such an example, the delivered infrastructure pays back (or offsets) the cost of issuing the federal reserve notes. Again, I don't see why--for specific infrastructure--it is necessary to take existing capital out of the economy, lock it into bonds, and require the government to play along with the theater of paying interest.

Respectfully you are arguing against a strawman.
The point is low borrowing costs help more projects have a positive npv.
Other comments have already pointed out that knowing the project returns would make this post interesting.

It's hard to know the return on a project so people say zero as a political point not a Econ or finance point

Thank you for giving us a row boat with one oar.


Interesting article. Thanks.

The private hurdle rate calculation itself is mistaken; it only accounts for the equity component of investments. A great percentage of investment is financed by debt, which has rates a lot less. The weighted average cost of capital would be DEBT_FractionXInterest_Rate_on_Debt + (1-Debt_Fraction)X Cost of equity.

this also ignores some other problems with this comparison that others have noted.

OK, it's late here and I am tired. So just that I understand the point easily: do you mean that T-bills crowd out productive private sector investments?

Is the government in question a currency issuer? The answer to that question vastly impacts the answer to your question.

Comments for this post are closed