Should government be spending more when real interest rates are low?

Let's look again at Brad DeLong on fiscal policy:

The U.S. Treasury can borrow for thirty years, taking all CPI risk onto its own books, and pay only 1.83% per year in interest?



It's not just that a greater amount of government investment meets the benefit-cost test when the government can borrow at 1.83% in inflation-proof bonds for thirty years, a whole bunch of tax postponements do as well. And so do a whole bunch of expanded social welfare programs. And so do a whole bunch of government issues of debt which are then invested in risky private ventures.

…the cost of borrowing for the government has fallen–the market value today of future cash tax flow earmarked for debt repayment has gone way, way up–therefore we should dedicate more future cash flow to debt repayment by borrowing more. There is no "but even." Expansionary fiscal policy is a good idea,

To explain why I view it differently, let's start with a parable from the private sector.  Managers in large corporations often are given project "hurdle rates" of twenty to thirty percent, not because their cost of capital is so high, but because of agency problems.  If the manager were given a hurdle rate of eight percent, he or she might go crazy with new projects and the company would too easily meet its ruin.  It then follows that changes in the actual cost of capital, within moderate ranges, don't so much influence private investment and in this second or third best sense shouldn't.  If the hurdle rate is thirty percent, and the cost of capital falls from ten to seven percent for a business, that maybe looks like a big change but actually it's a change in a largely non-binding constraint.

All that is standard and it has long been stressed by Joseph Stiglitz, among others. Furthermore in regressions the cost of capital often fares poorly in predicting private sector investment and that is one reason why.  I don't regard the above as "proven" but it does seem to be "likely."

Now let's turn back to the government.  Pre-crash, it already could borrow at low real rates of interest.  And there already were unexploited public sector projects with high potential rates of return.  But we don't want our government to go crazy, spending money without limit.  There are agency problems and a lot of the money ends up spent poorly.  We instead wish to impose a hurdle rate on government projects.  Let's say we trust our government more than shareholders trust private sector managers.  That could mean a government project hurdle rate of, say, fifteen percent, which would be below the typical private sector hurdle rate.  I'm being generous in that comparison of course; note that mistaken private sector projects get reversed more easily, through bankruptcy, than governmental mistakes do.

OK, there's a hurdle rate of fifteen percent and the cost of capital to government falls from three percent to one percent.  How much does it matter?  Should it matter?  The hurdle rate — the most binding constraint — hasn't changed.

There are many ways of complicating and modifying this model, and I am not saying that the correct "elasticity of projects," with regard to the real interest rate should be zero.  Still, I think this model explains why I am less impressed with the real interest rate changes than is Brad.

There is the separate — and very important issue — of covering for the worst-case scenarios.  This is stressed, correctly, in analyses of global warming but it matters in fiscal policy too.

One could make the quite different argument that we today have a better administration than average and that therefore the hurdle rate should be lower and government spending should be higher.  Now is not the time or place to evaluate such claims, but in terms of logic that to me is a stronger argument than citing the real interest rate effect.  It focuses on the most binding constraint, namely the hurdle rate stemming from the agency problems.  At the same time, it is an argument for higher spending with or without Keynesian factors or unemployed resources playing a major role.  It's the common sense point that a more competent manager should undertake more tasks.


If this is the case - if large private-sector firms act as if the interest rate were really thirty percent - doesn't that imply that changes in the real interest rate doesn't affect the private sector (much?) either?

I mean, presumably someone somewhere acting between the subordinate agencies seeing the hurdle rate and the bank offering the real rate must be making the judgment call. So what's different here?

As long as there is an idle resource somewhere out there, according to Brad DeKrugman, there is an opportunity for fiscal policy and more government. Those low interest rates are just a bonus.

In my world, when interest rates drop, hurdle rates drop as well. Still this model seems simplistic. Hurdle rates may be higher than cost of capital for new endeavors, but for existing operations, not so much. With state and local governments facing tremendous budget problems and high unemployment rates, i think it shouldn't be to hard to find ways to spend lots more without having to take on new stuff.

But aren't these low interest rates on Tresuries artificially low, due to Federal Reserve policy? The Fed is currently holding $776 billion of Treasuries.

In addition, the Fed holds $1.113 trillion of mortgage-backed securities, which also keeps interest rates on Treasuries artificially-low by reducing the crowding-out effect.

If the Fed sold all of these assets today, what would be the effect on Treasury rates?

how does your analysis apply to "tax postponements", which DeLong specifically suggests? a temporary cut to payroll or corporate income taxes shouldn't be affected by the agency/hurdle costs you focus on (unless you argue that the private sector will misspend tax cuts).

Frankly, Tyler, I don't know what you are talking about. Let us assume that my outstanding debt is $10 trillion and the average cost of funding it is 3% per year (in addition, I'm running a deficit and everyone knows that I'll be running a deficit for many years). Now changes in market's nominal interest rates decline and I may consider first whether to refinance my outstanding debt to take advantage of lower rates. Let us assume that the cost of doing it is zero so I do it and later today the average cost of funding my outstanding debt of $10 trillion is 2% per year. This debt relief reduces the expected deficit. At the end of today, I may consider whether to borrow some additional amount to finance new spending that indeed will increase the expected deficit. Let us assume that as result of that decline in nominal rates the relevant marginal cost of funding has also declined, but to take the decision I have to determine first how much I will be borrowing. Since the marginal cost of funding is an increasing function of the total debt, even if the curve has shifted downward, it is important to know how much I want to borrow (my flow demand for funds). I cannot assume that I can borrow any amount I want at a constant marginal cost and I know I should be worried how quickly marginal cost may increase with total debt. Thus, I have to assume that lenders are students of good professors of microeconomics.

Indeed, lenders will like to know what I'm going to do with the additional funds. Let us assume they are not worried about my management of the funds that I have already borrowed. Anyway, if I'm going to increase my spending in $X trillion (where X<1 or X<10% of the outstanding debt) in N years, lenders will like to know whether a bridge will be built (ignoring experiences like the reconstruction of the Bay Bridge or the "bridge to nowhere") or a transfer to old people will take place. Thus, I have to assume that lenders are not students of brilliant professors of macroeconomics. Now let us assume that I'm Al Capone. How is my Laffer curve? At what rate of taxation/extortion will I maximize my revenue? How can I guarantee my lenders that I will use my future revenue to pay back them? Will lenders be concerned that I can be overthrown by any of my children or by a younger Godfather? Under what conditions may lenders fear that I will kill them (sorry confiscate) them? To what extent does the risk premium I'm paying now reflect all these concerns? I better assume that lenders are students of good professors of public choice.

A P.S. to my previous post: RoR on public works is notoriously hard to measure because the government doesn't actually capture the return from a new freeway, it effectively hands it away to any person or business with a motor vehicle. But no one seriously disputes that the *societywide* RoR on most such projects (aside from bridges to nowhere) is quite high, even though it is not directly captured by the government and thus *looks* like a loss if you don't analyze in enough detail.

In addition to agency problems their is also the winners curse.

In my area this is a very real concern. Many people looked at the prospective project and may have passed. The fact that you did the project means you have the highest value in your analysis. But given that all analyses have uncertainties it is likely that you have a high value because your assumptions are wrong.

Consequently all projects should have a minimum return that is somewhat divorced from the cost of capital to cover for the chance of systematically picking bad projects.

This is even a bigger issue with the government who is not subject to the profitabliltiy pressures that constantly force you to refine your analysis methods. On any project that shows a high return the analysis is a likely to be wrong as it is likely to be a good idea.

Isn't this the same logic that leads to 95% LTV mortgages, over-leveraged LBOs, etc?
It isn't just the cost of borrowing but you do have to repay the principal. I don't agree that the hurdle rate is necessarily 15% or more - it seems to me they just has to be accretive to tax revenue net of the cost of borrowing, but I think very few government expenditures are plainly accretive to GDP.

I also question the facile elimination of inflation risk. I assume it's beleived that inflation drives tax revenues up by the same rate as the CPI but I wouldn't have confidence in that over a 30 year period.

Oh, and this statement appears misleading:

"The U.S. Treasury can borrow for thirty years, taking all CPI risk onto its own books, and pay only 1.83% per year in interest?"

The 1.83% is the rate for what they HAVE borrowed. We can't be sure they can borrow more at the same rate, can we?

"the price depends on the taxing power of the government"

yes, and my term for people who lend to the government for projects that can only be paid for by tax INCREASES rhymes with pay-holes.

Seems like an argument for a FICA holiday or at least elimination of FICA taxes on the first 25k of earnings. People may not benefit from new iffy projects but surely they benefit from having more money.

BTW wouldn't a FICA holiday be equivalent to individuals borrowing at the federal government rate. Of course there is a problem with women quitting taxed work when raise the tax rate latter pay back the loan.

Andrew: "We can't be sure they can borrow more at the same rate, can we? ,,, the %1.83 may not be sustainable."

I completely agree, Andrew. Any organization - corporate or government - needs to be extremely conservative in their capital budgeting process and use a very long term cost of capital.

we now have a considerable amount of new evidence on the Keynesian stimulus theory and the reality is clear: it works very rarely, and in most cases is highly counterproductive.

In its simplest form as understood by leftist politicians and many of the dozier members of the general public, Keynes' theory is nothing more nor less than a fallacy. It fails to recognize that the stimulus money has to come from somewhere, and that withdrawing it from circulation will have an opposing effect on jobs that may cancel out the Keynesian effect of the spending.

"One could make the quite different argument that we today have a better administration than average and that therefore the hurdle rate should be lower and government spending should be higher."

I would like to see someone take a crack at that argument. I am sure it would be highly amusing.


I don't think anyone disputes that. But consider that 90% of the time (more like 100% in the last 8 years) I hear the name Grover Norquist it is from people like you. In fact, I think in the last two years it has been 90% YOU.

Just because someone utters "starve-the-beast" as a concept, doesn't mean we are actually doing it. It's just laughable. Remember, my side's problem with Bush was the spending. If we couldn't do anything then, are we really holding up the show today?

1.83% for 30 years, no that can't change in less than 30 years.

You know what signals a studied, nuanced & open minded idea? Beginning with a typed throat-clearing. It's amazing how willing people are to tell you who they are.

Why does anyone take DeLong seriously? Any time you present counter evidence to his claims on his blog, he will delete the comment. No wonder he sides with the left; he runs his blog like Soviet Russia.

But seriously, has he never noticed that financial markets seem to reach "tipping points"? Sudden changes in heart? Should we assume that Treasuries are risk free no matter what, just like some assumed housing prices always rise? Does he assume that markets give "fair warning" to countries who borrow too much? How does he think crises arise?

DeLong believes in Keynesian style stimulus because he does not understand the role of credit in the economy. Sure, aggregate demand, if that's how you want to lump together as total spending on heterogeneous goods, has dropped. But how did total spending rise so far in the first place? CREDIT BUBBLE. I would like, for once, to see DeLong address head on how he thinks that the total spending levels achieved pre-recession were sustainable, given that they were based on a very clear credit bubble (look at a chart of revolving credit, mortgage loans/MEWs for evidence). People were taking out loans they could not possibly repay, based on the mistaken assumption that the value of their assets would rise. How does Brad DeLong think that is sustainable? He cites Say's Law as the reason to allow for a large stimulus. Say's law is based on a barter economy without credit, however. It is irrelevant to the discussion.

And why does he recommend that we try to sustain the status quo? Debt is debt. A country's citizens are ultimately responsible for its public debt. Sure a country can print money and an individual cannot. Does he support the destruction of savings value via printing money? How does a stimulus, without regard for how the money is spent, help our position and improve our standard of living? We'll see just how great the stimulus has worked in Q3 and Q4 while it is being withdrawn.

Here's a preview:

"Money well spent, idiot."

"But if only it had been bigger..."

Given the effort the Goldman's and Lehman's put in to satisfy the demand for toxic assets from their clients, so much so they created incentives to wrote mortgages to people who might as well be dead because there was nothing at all anyone was willing to invest in, why shouldn't the Federal government create lots of debt, bad or not, because investors are even more desperate for Federal Treasury debt they will almost pay the Treasury to borrow money.

And we must remember, the reason Bush embarked on creating several trillion in new debt in 2001 was to ensure the private sector has a good supply of Federal government debt to buy.

We have had about 10 tax cut to create debt and put money in investor pockets to invest more wisely than government, and it turns out that the best investment is in Federal government spending. Look at the interest rates!

The market is always rational, at least that's what the wise men like Milton Friedman tell us, so given the competition for Federal government debt, clearly there are no better investments possible than more Federal spending.

If there were good alternatives to invest in, then the market would drive up Federal government debt's cost to reflect the much wiser non-government investing options.

But the private sector can't come up with any. The market says so, and that is as good as Milton Friedman saying the best investment today is Federal government spending.

@mulp -- unfortunately, when Americans are overpaid, you are correct that there are few good prospects for entreprenuer / investors to invest in, in America.

@al, thanks for a nice note, reminding me of a song ... Yes, we have no bananas, we have no bananas, today.

@Dave, while you have some good points, you really should understand DeLong-Krugman's key goal -- maximum sustainable employment. I recall getting my 2004 comment about how 4-5% unemployment under Clinton boom years was actually "over-employment", and not sustainable, and thus not fair to contrast with Bush. The gov't will spend its stimulus like a pack of banana eating monkey idiots.

@Ryan, how do we know the "most effective" way to boost spending? Starting with some 5 year base case of assumed spending, compare policies:

a) spending 5% more in year 1, (raising long term GDP &spending), 3% in year 2, and 1% in year 3 when the loan repayments kick in to reduce spending as a % of GDP, but at a higher GDP than base case.
1% more in years 4 & 5.

b) spending 10% more in year 1, but lowering long term GDP by crowding out private sector and increasing long term deficit and uncertainty, 3% more spending in year 2, -1% in year 3, -2% in years 4 & 5.

Before specifying which of a) or b) is Gov't stimulus or Tax Loans, as compared to the base case, do we have a way of saying which of a or b or the base case is the most efficient?
It's actually a really good issue, but obviously the policy advocated leads to bias in the presentation of the future alternatives.

And with overpaid Americans, as compared to E. Europeans (like my colleagues in Slovakia), or India Indians or Chinese in China (or SE Asia), the "true" base case is worse than any politician could get elected as claiming.

Solving gov't pensions, like switching them to Social Security only, is currently one of the most important long term issues that needs to be solved.
Gov't workers are the MOST overpaid, so real taxpayers need to change that before America is too much like Greece to stop a bigger crisis.

To me, the real interest rate is an argument that we should be refinancing all that bubble debt. Stop throwing all that money away on interest so that banks can it on it.

Comments for this post are closed