Low interest rates vs. high risk premium

by on May 30, 2012 at 6:57 am in Economics | Permalink

It is often claimed that the governments of the United States, the UK, and Germany should spend more money because they can borrow at low rates, thus raising the present expected return on the investment considered as a whole.

Maybe, but keep in mind that the interest rates on quality government debt are down, in part, because the risk premium is up.  Non-governmental investments are perceived as riskier.  It is also possible that governmental outputs are perceived as riskier, as those outputs will be evaluated by consumers.  Note that Kenneth Arrow’s “the government can spread around the financial risk” point does not eliminate this more fundamental risk, namely the risk associated with the quality of government output, just as there is a risk associated with the quality of private sector output.  Michael Jensen made this point in 1972.

You might think the government investments are “low hanging fruit” in terms of quality.  Maybe yes, maybe no, but the low real interest rate doesn’t signal that, rather it signals merely that people expect to be repaid.

In this argument for more government investment, the notion of government investments as low hanging fruit is doing a lot of the work.

1 Bill May 30, 2012 at 7:17 am

To summarize:

Maybe, but keep in mind, maybe yes, maybe no.

On the other hand, yes,


On the other hand, no.


2 david May 30, 2012 at 8:43 am

He-ah in the Amazonian rainforest lives the rare one-handed economist…

3 Peter Gordon May 30, 2012 at 7:29 am

What is the risk associated with high-speed rail? The odds are very high that it will be another failed mega-project. No matter who funds it, the project is high-risk. Paying for it via Fed policy-engineered low rates only masks the problem. I expect there are many items in the federal budget that are similar.

4 Doc Merlin May 30, 2012 at 11:26 am

Exactly! Government rates are low, not because of actual project risk, but because they can take money by force. This means that the risk is underpriced, which results in market distortion.

(Did I just say government was a market failure? Yes I did)

5 msgkings May 30, 2012 at 12:04 pm

And we’re all shocked, Doc.

So…market failures DO exist!

6 Willitts May 30, 2012 at 1:41 pm

Of course market failures exist. The question is whether government fixes the problem or makes it worse. In this particular case, government is distorting market signals which prevent the market from correcting. Only in the fantasy world of theoretical gibberish do these measures work. We are wasting resources in fruitless attempts.

Another question is whether there are decentralized methods of addressing market failures that don’t involve government intervention. Government never permits these to operate.

7 Firat Uenlue May 30, 2012 at 7:34 am

Unfortunately the idea that there is a lot of low-hanging fruit which could be harvested if only the government went into deficit to finance it, what these people somehow seem to miss is that the money is already there but instead we choose to spend it on everything but infrastructure. Then, when all taxes (and more) are spent on social welfare, military etc. people start to wonder why no money is left for investment and justify deficit spending via this route. Absurd.

8 Greg G May 30, 2012 at 7:35 am

I, for one, find it refreshing that Tyler often really considers different ideas rather than automatically parroting the current libertarian talking points. If the latter is want you want Bill (apologies if I’m assuming too much) there are plenty of blogs for that.

9 Bill May 30, 2012 at 7:53 am

Greg, I think Tyler’s post makes little sense.

To parse it:

1. “….they can borrow at low rates, thus raising the present expected return on the investment considered as a whole.” OK, stop there. If something raises the “expected present expected return”, by definition, it is including the discounted value of present returns and future returns to present value. Finance 101.

2. “Maybe, but keep in mind that the interest rates on quality government debt are down, in part, because the risk premium is up.” DAH, it is the risk premium of PRIVATE investment that is up, relative to public investment, not the risk premium of government. This means that the market is telling you that you would be better off, at the margin, going to where the market is telling you to go to use your money, PARTICULARLY when aggregate private demand is low, and you are in a liquidity trap, that is, low, and even negative real rates aren’t doing a thing.

3. “It is also possible that governmental outputs are perceived as riskier, as those outputs will be evaluated by consumers.” If government outputs are perceived as riskier, that is inconsistent with the initial premise that “they can borrow at low rates, thus raising the present expected return on the investment considered as a whole.” What Tyler is really doing with this statement is going back and saying: “Well, those things that government could do, which are better alternatives than what the private sector could do because the private sector is stuck–those things that I told you we could measure as being of better value, we really can’t measure. Trust me.” Whoa folks. Who said anyone is throwing out cost/benefit analysis or abandoning yardsticks on what is a good project and what is not: OR, to think about it a different way: IF we know that government will be having to purchase things in the future (schools, roads, replacement of military equipment), why shouldn’t we purchase now rather than later? This may also be a Finance 101 question as well: buy now when interest is low and costs are low, or buy later when rates are high and costs are high.

4. As to whether this is balanced: ask yourself: were the premises changed during the argument, were there assumptions made that were unstated as to the value of a government expenditure, does the present value calculation make sense.

You be the judge.

10 Bill May 30, 2012 at 8:04 am

Greg, If you want to look at the paper or discussion on this topic of spending now at the zero bound, look up the Summers/deLong presentation summarized here: http://delong.typepad.com/sdj/ The paper discusses why in normal circumstances, expansionary fiscal policy would be offset by the Fed raising interest rates, but in a situation we have here–even if you assumed very low multipliers, time discount, etc.–it makes sense now.

Do the math. Read the paper.

Or, you can, on the one hand, say yes, and on the other hand, say no.

11 Greg G May 30, 2012 at 8:22 am


I was assuming too much. Thanks for the thoughtful reply.

12 Neal May 30, 2012 at 8:00 am


-Scott Sumner

13 Doc Merlin May 30, 2012 at 10:59 am


14 Michael G Heller May 30, 2012 at 8:03 am

There’s an article in Financial Times markets section today reporting very big fund managers urging less austerity and more spending in countries where debt is cheap. They promise to remain steady in return for government backloading commitments. But whose to say what is the magic number at which they will start punishing those same countries for their debt ? Either way what governments need to do is to reduce their vulnerability and prepare public finances for rocky times. Governments should not listen to investors who want to safety in growth havens by promising not to attack spending plans… yet.

15 Rahul May 30, 2012 at 8:38 am

It is also possible that governmental outputs are perceived as riskier, as those outputs will be evaluated by consumers.

Why would people lend at low rates to an undertaking whose outputs they perceived as risky?

16 Norman Pfyster May 30, 2012 at 8:55 am

Because, unlike private enterprises, the input (revenue to pay creditors) has no relation to the output.

17 Bill May 30, 2012 at 9:00 am

Norm, If that is true, then you are saying that, at the margin, government debt does not compete with private debt. The government could just take a stack of money and burn it and it would have no effect on private or public output, nor would it raise interest rates.

18 Salem May 30, 2012 at 8:57 am

Because bondholders do not expect to be paid back by the outputs. They expect to be paid back by taxes.

Example: government proposes building a new bridge, free to use, cost $1bn, expected lifespan 100 years. It finances the bridge entirely through debt.

There is no contradiction in saying this is a very safe investment for a bondholder, and a very risky investment for a taxpayer.

19 Rahul May 30, 2012 at 9:02 am

And yet, even such a risky, incompetent, unaccountable government is better off financing the bridge in a low-interest year than a high-interest year? Maybe the bonds get paid via taxes either way, but less taxes the better?

20 Willitts May 30, 2012 at 1:49 pm

The lower interest rate makes more projects admissible, but it does so only by ignoring that fact that future discount rates will be higher than the rates used today. Another problem is that the government’s asset beta has risen, and therefore the weighted average cost of capital may be higher. The reduction in the risk free rate lowers WACC. The increase in the market risk premium raises WACC. So the nation’s asset beta controls the outcome.

Accelerating the replacement of existing infrastructure destroys assets which have not reached their salvage value and are still as productive as they ever were. Government is acutely aware of its income statement, but not its balance sheet.

Building a project that doesn’t exist faces the prospects of cannibalizing existing assets as well as raising input costs for private projects. Further, we engage resources in the construction of long-lived assets. This builds human capital, but for skills that the economy might not need when the projects are done. Long-lived assets put their resources out of work the moment they are completed. This is exactly how we got so much unemployment, and this is not how we solve the problem. It is a phony cure.

21 Bill May 30, 2012 at 9:03 am

You are making an assumption that you do not need to do a cost/benefit analysis. Moreover, if you were simply burning cash, as you imply, you would raise interest rates, making private investment a more attractive alternative.

Why not, instead, focus on something realistic: like moving forward purchases you would make in the future anyway? That’s not a strawman, like your bridge example that is not justified by cost/benefit.

22 The Anti-Gnostic May 30, 2012 at 9:37 am

Why not, instead, focus on something realistic: like moving forward purchases you would make in the future anyway?

That’s how the schlep in the credit card doom loop thinks too, only to find he’s that much poorer in the future. No different when the State does it, either by debt or money-printing. This is the biggest house of cards since 2008.

23 Bill May 30, 2012 at 9:54 am

Anti, That’s also how a schleppy futures market works, too.

24 TallDave May 30, 2012 at 10:08 am

A better question is: why would people lend at low rates to a notoriously poor “investor?”

And the answer is guns. This investor can just seize more revenue.

As I said the other day, there are all kinds of worthy public goods, but relatively few of our tax dollars are spent on them. And the more the gov’t spends, the less effective that spending is likely to be.

25 Doc Merlin May 30, 2012 at 11:00 am

Public spending crowds out public goods spending.

26 steve May 30, 2012 at 9:02 am

risk premia are by definition premia over the risk free rate. how can a high premium push down the risk free rate?

27 Willitts May 30, 2012 at 1:59 pm

You are technically correct, but are forgetting the dynamics. The risk free rate, the risk premium, and the project beta determine the cost of capital.

The risk free rate is low, but artificially so. It would be a mistake to conclude that a large project can be financed at such low rates without discount rates rising in the future.

The passage asserts the risk premium is high. Let’s take that at face value.

What has happened to the level of systematic risk, i.e. the beta? Many liabilities grow in value when interest rates fall because the discount rate has dropped. The liabilities being underfunded means there is increased pressure to raise returns at the same time that risk tolerance has declined. If we accept more risk to improve returns, then Beta rises and so does the cost of capital.

In other words, while interest rates are historically low, the observation of low investment, low home purchases, low output, low loans, indicates that perceived systematic risk is VERY high, which makes the cost of capital very high.

28 Nonsense May 30, 2012 at 9:13 am

Both absolute and relative risk premiums across the board are hitting new decade lows, particularly as measured by credit spreads on fixed assets. So I’m not sure what this blog post is attempting to convey.

29 TallDave May 30, 2012 at 10:04 am

Good points.

Remember, Greece got into current their mess by increasing spending 10% per year when credit was cheap in the early euro era. They don’t seem to be reaping a lot of growth from their “investment.”

30 Michael May 30, 2012 at 10:23 am

There are many interesting questions we’d like the answer to (eg. what are these real low ineterest rates truly signaling?) Understanding the mechanism that’s driving these results is undoubtedly important for understanding how the global economy works or how investors perceive the future to look.

On the other hand, we don’t necessarily need a clear answer to those questions in order to do a cost-benefit analysis of doing something like upgrading the electrical grid at the current interest rate. Moreover, if the project in mind is something that should be done eventually, no matter what (and updating the power grid is likely one such project), then it’s even simpler as it’s a matter of timing. With something like that, the finance 101 questions concering ROI or whatever aren’t all that important. It will eventually have to get done. The question then is, is it better to do it now or later? That a function of when you think interest rates will be lowest, when it’ll have the least impact in terms of crowding out private investment and when the effects on the private labor market are least detrimental.

It seems to me that undertaking such projects now looks like a pretty opportune time to do so. I’m aware and sympathetic to the notion that many people would prefer they be financed by cutting spending from elsewhere rather than through new debt. One may think that’s the first-best option. But if by politics or whatever, that’s not entirely feasible, financing such projects with new debt may be the best feasible option (be it the second, third, or fourth best option). At what point does the best available option include NOT undertaking such projects now? Whatever that point is, I don’t think we’re in that situation. Nor do I think answering the riddle of what precisely is being signaled is a necessary condition for deciding whether or not to do it. It may indeed be helpful to know what precisely these rates mean when doing the cost-benefit analysis of new infrastructure projects, but those answers aren’t a necessary condition to answering the more basic questions of, “Will these things need to be done at some point regardless? Is doing them now likely to be cheaper and less distortionary than if we wait?”

That is, I think all this “one the one hand…on the other hand,” theorizing stuff does more harm than good in the sense that it’s distracting us from answering the more relevant question. Go ahead and work on both simultaneously, but don’t get fooled into thinking that answering one set of questions is required before one can answer the others. Helpful, yes, but required? I don’t believe so.

31 Jason May 30, 2012 at 11:11 am

“Motivated Skepticism in the Evaluation of Political Beliefs”, Taber, Charles S.; Lodge, Milton, American Journal of Political Science, Volume 50, Number 3, July 2006 , pp. 755-769(15).


32 Popeye May 30, 2012 at 12:46 pm


33 Bill May 30, 2012 at 9:26 pm


34 Meegs May 30, 2012 at 11:16 am

Another reason rates are low:

Inflation. Or lack thereof.

What happens to rates if the FED commits to higher inflation?

35 Brian Donohue May 30, 2012 at 12:53 pm

Lack of inflation? Since December 2008, the US dollar has lost 10% of its value. 3.5 years- yeah that feels pretty modest (but I’m a child of the ’70s.) But since the return on savings since then has been around 1% total over 3.5 years, the value of savings has eroded 9% over this period. Assuming $2 trillion parked in ‘anti-social’ savings over this period, that’s $180 billion in covert taxation due to punishing savers. Keynes would be proud.

“Real” rates are easily at all-time lows right now. But don’t believe me- see what you can get in TIPS.

36 Rahul May 30, 2012 at 11:29 am

What percent of American debt gets bought by non-American investors, anyone know?

Is it merely Americans trusting their Government or non-Americans or both?

37 The Anti-Gnostic May 30, 2012 at 12:26 pm

“Investors.” How quaint: Ma Kettle patriotically loaning money to the gubmint at negative interest.

Any investors are surely just parking money in what they presume will always be a highly liquid market. Until it isn’t. Also, how much of the debt is US and foreign central bank purchases? Aren’t the Fed’s primary dealers obligated to purchase a portion as well? USG debt is a huge footprint in the capital markets

38 Willitts May 30, 2012 at 1:50 pm

About half of American debt is held by Americans (including government), and half by foreigners.

39 derek May 30, 2012 at 12:51 pm

This seems predicated on the idea that this is a cyclical downturn where you buy the dips.

I would posit that it is not and the extremely low interest rates are a reflection of that fact. Today you almost have to pay the Swiss to hold their bonds.

The situations is this. Global debt increased by clos to 2.5 times since 2002 from$ 80 trillion to$ 210 trillion. Too much, unaffordable. What will happen is that a portion of this debt is going to be written off. Let’s say in the end there will be$ 160 trillion. $50 trillion is going to be lost. This week it is spanish banks and real estate debt being valued far below par. Money is fleeing into something, anything that participants feel will have some value in 30 days.

There are smart ways to handle this. Most important is to maintain the perceived value of your offerings. I doubt borrowing money to buy granny’s hip replacement will do.

40 Hideous May 30, 2012 at 3:13 pm

The Federal government “borrows” at low rates because it is simply printing 3/5 of the funds allegedly “borrowed.” This shill-bidding in Federal debt auctions drives down the rate; most of the remaining debt buyers are there under compulsion (like banks and insurers whose regulators require them to buy Treasury securities) so they just have to eat the artificially-low rates.

As my friend alluded to above, there is also a “flight to safety” problem pushing private funds into non-yielding gov’t paper, but that too is partly or largely forced by government. Federal authorities have intervened to prop up the values of mortgage bonds and so-forth in the interests of big banks. Everyone knows those are far riskier than the government says, but since those are propped up by fiat, they are poor investments (no real economic value, only fickle political value)– so loss-averse investors avoid them and end up in government paper by default.

41 The Anti-Gnostic May 31, 2012 at 2:13 pm

What a sweet deal. You run a trillion dollar deficit and instead of all your checks getting returned NSF, the bank just cashes them. Then, when you issue a bunch of IOU’s to “cover” your shortfall, the bank obligingly purchases them at rates lower than what your own grandmother would charge you. No wonder central bankers are regarded with such awe and reverence.

42 John May 30, 2012 at 8:50 pm

Readers puzzled by this post should read Noah Smith’s commentary at

43 Woj May 30, 2012 at 11:50 pm

From an MMT/MMR/Post-K perspective, sovereign currency issuers do not strictly require financing in order to spend their own currency. Treasuries are therefore used primarily to manage interest rates and transfer interest income to the private sector. Edward Harrison explains that “Long-term interest rates are a series of future short-term rates.” (http://bubblesandbusts.blogspot.com/2012/05/permanent-zero-record-low-treasury.html)

Current low rates are therefore reflective of expectations that the Federal Reserve target rate will be maintained at very low levels for many years to come. Using the low rates as a reasoning for investment is flawed. The government’s decision to invest should be about the potential return of a specific opportunity versus other current and future opportunities, int he context of how much total investment is desired.

44 stan June 1, 2012 at 8:56 pm

Hideous has it right. The present interest rates are manipulated.

The essence of the argument that govt should borrow because rates are low fails to account for the fact that they are only low because the Fed is using monopoly money to bid them down. The Fed could bid rates to zero forever under this plan. Of course, inflation would destroy the country, but the govt can insure zero interest rates any time it wants by simply printing (as it now) rather than genuine borrowing.

You have to wonder a little about folks who push an argument that says govt should borrow now because it has found a sucker willing to lend at very low rates. Who is this sucker? The govt.

45 stan June 2, 2012 at 12:35 pm

Restated — government can’t borrow at these low rates. When the Fed buys the debt, that’s not borrowing. That’s just printing money. Government ‘borrowing’ requires someone willing to lend money to it. No one willing. So the Fed ‘buys’ it with newly printed cash.

Only an economist could look at a market where no one is willing to lend and see an argument for borrowing even more from those unwilling lenders.

Comments on this entry are closed.

Previous post:

Next post: