Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Thursday, April 28, 2016

On the want of U.S. government debt

In a recent article, Narayana Kocherlakota lays out the case for why, under present conditions, the U.S. government should be issuing more debt, using the proceeds to cut taxes, finance infrastructure spending, or both. It's a policy that many economists, including yours truly, have been advocating for some time. And while I generally support the policy, I thought it would be useful, nevertheless, to reflect on some possible counterarguments. It's not a slam dunk case, one way or the other, I think.
Kocherlakota does a good job explaining why a deficit-financed tax cut, or deficit-finance infrastructure spending is a good idea. I want to make it clear that the argument in favor of the policy hinges critically on the presumption that we can rely on Congress to manage the public debt over time in a responsible manner. Let's accept this assumption, provisionally at least, in order to understand the economic argument. I will come back to the political argument later.

While the debt-to-GDP ratio (D/Y) is presently high by historical standards, it's not unmanageable. The key is not the D/Y itself, but its trajectory over time. Clearly, D/Y cannot grow forever. And fortunately, market signals are available to monitor how the public perceives the likely path for D/Y over time. These market signals are: (1) the yields on U.S. treasury debt (at various maturities), and (2) inflation and inflation expectations. So what are these market signals telling us? The yield on U.S. treasuries is presently very low. Both inflation and inflation expectations are presently running below the Fed's 2% target and have done so for years now. So far, so good.

The large increase in D/Y since 2008 together with plummeting yields and low inflation may seem puzzling, but it's not really. Usually, a bad event that triggers a large increase in the public debt also triggers higher bond yields and the prospect of inflation. We can expect this to be the case in any experiment where the supply of debt increases in the face of a stable (or diminished) demand for the debt that is being issued. Think Zimbabwe or Venezuela.

But the U.S. is not Zimbabwe or Venezuela, or the Weimar Republic, for that matter. Rightly or wrongly, the U.S. treasury security is viewed by investors around the world as a safe haven asset. So when the financial crises hit in the U.S. and Europe 2008-10, investors moved en masse into U.S. treasuries (and other sovereign debt instruments viewed to be relatively safe). In short, while the supply of U.S. debt spiked up, the demand for U.S. debt increased by even more. We can infer this from the behavior of bond yields, which went down (the price of debt went up) at the time.

So the economic argument is simple. The U.S. government can presently borrow at essentially zero interest (more or less) even 10 years out and more. This effectively gives the fiscal authority the ability to print money (low-interest debt), so there's no need to rely on the Fed. To the extent that domestic real economic activity is still not firing on all cylinders, why not offer temporary tax cuts to stimulate demand? Why not re-build that crumbling infrastructure, putting people to work, all financed at zero-interest? It sounds like a no-brainer.

Alright, now for a couple of counterarguments, one economic and one political.

An economic argument against temporarily increasing the public debt further (and indeed, taking measures to reduce it) could be made on the basis of the Triffin Dilemma. The economist Robert Triffin noted back in the early 1960s that world reserve currency/debt status is a double-edged sword. On the one hand, it's great that the U.S. can just print paper that is coveted around the globe. If foreigners are willing to export their goods and services to us, expecting only paper in return, then we are extracting wealth from the rest of the world (in exchange for what ever financial service our paper is providing them).

One implication this power, if exercised, is that the world reserve currency issuer is likely to run persistent trade deficits. Triffin worried that the huge amount of U.S. currency held by foreigners exposed the U.S. to foreign risks. What might happen, for example, if foreigners suddenly decided they no longer wanted to hold USD or USTs? This could result in a sudden and dramatic change in the exchange rate, leading to domestic inflation and sharply higher bond yields.

There is also the trade-related argument that persistent trade deficits kill domestic industries and domestic employment. After all, if we can make the rest of the world work for us in exchange for paper, where is the need for us to work at all? The implied boom in domestic leisure consumption sounds good theoretically. But of course, in reality, the gains are not evenly shared. The rich gain by purchasing cheaper foreign goods. The poor are out of their jobs.

A political argument against more government debt could be made by challenging the assumption that it will be managed responsibly. This "we can't trust future politicians to do the right thing" argument is (sadly) not without empirical merit. I am reminded of the following quip by P.J. O'Rourke,
"The Democrats are the party that says government will make you smarter, taller, richer, and remove the crabgrass on your lawn. The Republicans are the party that says government doesn't work and then they get elected and prove it."
I can't help but note a certain irony here. There seems to be a strong presumption among people (Americans in particular) that the government should run its finances in the manner of a household. Economic theory is quite clear that this sentiment, however noble, is just plain wrong. The irony is that to the extent that this sentiment finds its way to being represented in Congress, it proves to be a very valuable "anchoring" device for the fiscal authority.

That is, I sometimes wonder whether US treasury debt is valued around the world the way it is precisely because it is known that Congress is impregnated with a large number of genetic "debt-ceiling" algorithms. It may not be an ideal situation from the perspective of pure economic theory, but then again, it's not hard to think of worse scenarios.


  1. IMHO the first (economic) counter argument wouldn't be valid as long as there was idle capacity & people available for employment. Moreover, under the current "liquidity trap", no presumable interest rate hike would crowd out private demand, at least most of it.
    As for the "political" one... well, to assume that Tea Party & other Congress republicans' varieties are the only fiscally "responsible" politicians, sounds a little far-fetched. Worse, much of their rhetoric sounds pretty irrational, as too much austerity lowers long run growth prospects through negative expectations, poor private investment, & hence less taxable income thereafter.
    So your political argument implies that all foreigners are as crazy as the GOP's current superstructure, in believing that "sequester", and other regrettable methods to reduce expenditures are the best way to long term fiscal reliability.

  2. The unintended consequences of deficit spending
    Comment on David Andolfatto on ‘On the want of U.S. government debt’

    You conclude: “There seems to be a strong presumption among people (Americans in particular) that the government should run its finances in the manner of a household. Economic theory is quite clear that this sentiment, however noble, is just plain wrong.”

    What is curious about the deficit discussion since Keynes is that profit is entirely left out of the picture. The ultimate reason is that Keynes got profit theory wrong and After-Keynesians simply parroted Keynes’s blunder. As a result conventional profit theories are provably false until this day. As the Palgrave Dictionary summarizes: “A satisfactory theory of profits is still elusive.” (Desai). This is not exactly a great scientific achievement.*

    The correct profit equation in the most elementary case of the pure consumption economy is given by Qm=C-Yw. Legend Qm: monetary profit, C: consumption expenditures, Yw: wage income. Monetary profit of the business sector is equal to the deficit spending of the household sector.

    The profit equation for the investment economy reads: Qm = Yd+I-Sm.** Legend Qm: monetary profit, Yd: distributed profit, Sm: monetary saving, I: investment expenditure.

    When we leave this part unaltered for the moment and add the government sector then we have: the increase of overall monetary profit of the business sector is equal to the deficit of the government sector.

    And here we arrive at the irony/absurdity of the balanced budget discussion. Normally, the friends-of-the-workers argue in favor of public deficit spending in order to push employment. This has the unintended consequence that overall profits increase one-to-one with the growth of public debt. On the other hand, the friends-of-the-capitalists argue in favor of deficit reduction in order to balance the government’s budget. This has the unintended consequence that overall profits decrease. In full ignorance of the profit effect on the distribution of income and wealth BOTH sides argue ultimately AGAINST their own interest. Obviously, this constitutes the worst violation of the foundational principles of economics, i.e. of self-interest and utility maximization.

    Egmont Kakarot-Handtke

    * See ‘The Profit Theory is False Since Adam Smith. What About the True Distribution Theory?’

    ** See ‘The Three Fatal Mistakes of Yesterday Economics: Profit, I=S, Employment’ eq. (18)

    1. "... BOTH sides argue ultimately AGAINST their own interest."

      Only if you believe the monetary profit of the business sector is the exclusive domain of the friends-of-capitalists.

  3. "...if we can make the rest of the world work for us in exchange for paper, where is the need for us to work at all? The implied boom in domestic leisure consumption sounds good theoretically. But of course, in reality, the gains are not evenly shared. The rich gain by purchasing cheaper foreign goods. The poor are out of their jobs."

    Excellent. +1000
    If currencies were valued in a way that the balance of payments between countries was well, more balanced, then the dollar would be lower, domestic production/employment/income would be higher, the tax take would be higher to plug the budget deficit, and the demand for our safe assets would drop and interest rates would rise and then *domestic* savers would be rewarded. Sounds like a win-win-win to me. When do we get some kind of new Bretton Woods agreement? SDR's, import/export certificates, whatever, bring them on. The status quo can't go on forever, and I don't want to experience the unwind of the past 20 years.

  4. Fed St. Louis cannot handle the thought that funding is independent of spending?

  5. While those economically illiterate politicians are fully entitled to take strictly POLITICAL decisions, like what % of GDP is allocated to public spending, the idea that they have to qualifications to decide how large the debt should be is just laughable. At some time in the future, hopefully the latter decision will be taken just be economists.

    1. Well, if the tax and spend decisions are made by the legislative branch, this doesn't leave the treasury division much room to maneuver, does it?

    2. There is a cyclical component to government debt, because it offers its balance sheet to insure us against macroeconomic fluctuations. If the treasury unloaded this risk with suitable macro securities, then long-term trends in debt would be more apparent, and this would increase the accountability of legislators for their debt legacy. As it stands now, we spend a lot of time debating the comparative economic legacy of different administrations, because the evidence isn't very clear.

      But yes, the treasury can't make such a large structural change unilaterally.

  6. Leaving aside for the moment how the money would be used, including the important public choice implications, I question the source of the demand for U.S. debt right now. I recall your paper, I think with Prof. Williamson, about high-quality debt & the relative paucity of it. I don't have it at my fingertips, but the paper indicated that the cause of the collapse in the money market was a lack of high-quality assets used to back overnight lending, yes? Empirically, I think the argument was made that one of the few high-quality assets right now, as in 2008, is U.S. Treasury debt.

    Recently I've seen some data that China & Japan are trimming their holdings of U.S. Treasury debt, but that hasn't really moved the needle on yields. Clearly there's demand for Treasuries. I think the demand is related to Basel (I,II, and III) accords for banks, because Treasuries can be held as an asset with no risk capital requirement. Other assets carry (relatively) high capital requirements for yields that may not be that much higher than Treasury yields, so bankers like Treasuries.

    If the regulatory story is true, these conditions could persist for a long time. After all, the 0% capital requirement has been with us at least since Basel I, which the G-10 starting enforcing in 1992. But if it is true, aren't there concerns over impediments to other types of lending that could limit growth in the private market?

    1. My impression is that the "regulatory demand" for high grade sovereign debt is significant. Could you elaborate on what you mean by your concluding question?

    2. What I'm getting at is that the spread between a 10-year Treasury (1.85%) and a 10-year commercial loan (3.5 - 5.5%) may not be enough to get a bank to tie up risk capital in anything but the least risky commercial loans.

      Another example is the spread between, say, 30-year Treasuries (2.68%) and 30-year FRM (3.64% average). A residential mortgage requires 50% or 100% depending on its risk profile. If the bank were to have to hold the mortgage, that spread might not justify the equity it needs to tie up. And banks can't immediately sell the mortgage, it takes a few months to build enough inventory to package up into a MBS of some type.

      I'm just thinking off the top of my head, so in the final analysis the presence of other investors might mitigate these lending problems, or they might be small in the first place.

    3. Yes, but are suggesting that increasing the supply of UST will alleviate this problem, or make it worse?

  7. My question is how much is US monetary policy now global monetary policy? is the combination of open swaps with foreign central banks combined with the new accounting rules on safe assets at foreign banks made the Fed a lender of last resort for the Eurodollar market.

    There is a very large demand for dollar denominated safe assets at banks to cover the reserve requirements of dollar denominated foreign loans of foreign dollar deposits.

    As long as the US is the only major reserve currency and the swap lines with foreign central banks exist I don't see how the Fed can ever shrink its balance sheet anytime soon. The dearth of safe assets was largely an international phenomena and much of the financial crises was a liquidity crises in the Eurodollar market. The biggest reason the Fed had any room to raise interest rates was the cratering of the Oil price that reversed the flow of safe assets and thus provided a temporary source of safe dollar assets, however this could change if Oil prices rise enough or the major petrostates rebalance their trade flows to once again accumulate dollars.

    The Fed is now the worlds central bank, it is explicitly linked to much of the worlds financial system and the demand for safe assets will be driven by the need for safe assets to meet the reserve requirements of foreign banks who drove the creation of dollars in international investment (foreign banks creating dollar denominated liabilities and assets by lending to foreigners). The banks of the world now must meet more stringent reserve requirements and they need the assets to do so.

    This is the part that is missing. The demand for US debt is international since so many of the worlds loans are dollar denominated and to finance these transactions they need to hold a certain amount of safe dollar assets (although there are further ways to do this by holding non-dollar assets that are combined with dollar currency hedges). This was all a whole lot easier when the easiest way to do this was to rely on the massive surplus of petrodollar deposits.

    The irony is the whole point of monetary policy is to restrict the amount of safe assets to the extent that the price is so unattractive that savers are forced to actually find someone else to lend to. If they can't find anyone then that may be a structural issue and the government will have to just suck it up and do its best to take the money being offered to it.

  8. I agree with the basic thrust of the above article, but not with this passage: “Why not re-build that crumbling infrastructure, putting people to work, all financed at zero-interest? It sounds like a no-brainer.”

    The reason governments can borrow at exceptionally low rates is that governments, first, can simply print money to repay their debts, or second, they can grab any amount of money they like off taxpayers to repay their debts.

    A private contractor who borrows to create infrastructure has no such luxuries. Thus anyone lending to such contractors will want a much higher rate of interest. I suggest the latter rate is the “commercially realistic” rate for infrastructure.

    If government is going to borrow to fund infrastructure, those buying relevant bonds should be exposed to the latter sort of risks (as they were in the case of the British – French Channel tunnel) and government should have to pay the relevant rate of interest.

    An alternative, advocated by Milton Friedman and Warren Mosler is for government to issue base money, but no interest yielding bonds at all. Though those using infrastructure should still pay an imputed rate of interest.