Why Not Arbitrage TIPS and Treasuries?

Larry Summers has a really interesting piece in today’s Financial Times (“Look beyond interest rates to get out of the gloom”), advocating that safe-haven governments (like the US, Germany and the UK) which are now able to borrow at close to zero rates of interest, which adjusted for expected inflation, amount to negative rates. Given such low borrowing costs, Summers argues, governments should be borrowing like crazy to finance any investment that promises even a marginally positive real return, even apart from Keynesian stimulative effects. This is not a new idea, countercyclical public works spending has often been advocated even by orthodox anti-Keynesians as nothing more than sensible budgetary policy, borrowing when the cost of borrowing is cheap and hiring factors of production in excess supply at discounted prices, to finance long-term investment projects. If there is a Keynesian effect on top of that, so much the better, but the rationale for doing so doesn’t depend on the existence of a positive multiplier effect.

But Summers’s argument takes this argument a step further, because as he presents it, the case for doing so is almost akin to engaging in an arbitrage transaction.

As my fellow Harvard economist Martin Feldstein has pointed out, this principle applies to accelerating replacement cycles for military supplies. Similarly, government decisions to issue debt and then buy space that is currently being leased will improve the government’s financial position. That is, as long as the interest rate on debt is less than the ratio of rents to building values, a condition almost certain to be met in a world of government borrowing rates of less than 2 per cent.

These examples are the place to begin because they involve what is in effect an arbitrage, whereby the government uses its credit to deliver essentially the same bundle of services at a lower cost. It would be amazing if there were not many public investment projects with certain equivalent real returns well above zero. Consider a $1 project that yielded even a permanent 4 cents a year in real terms increment to GDP by expanding the economy’s capacity or its ability to innovate. Depending on where it was undertaken, this project would yield at least 1 cent a year in government revenue. At any real interest rate below 1 per cent, the project pays for itself even before taking into account any Keynesian effects.

Now one blogger (Tea With FT) commenting on Summers’s piece found grounds to quibble about whether the rates governments pay on their debt are “free market rates,” because banking regulations allow banks to reduce their capital requirements by holding government debt but must increase their capital requirements as they increase their holdings of private debt.

Lawrence Summers is just another economist fooled by looking only at the nominal low interest rates for government debt of some “infallible” sovereigns, “Look beyond the interest rates to get out of the gloom“. Those interest rates do not reflect real free market rates, but the rates after the subsidies given to much government borrowing implicit in requiring the banks to have much less capital for that than for other type of lending.

If the capital requirements for banks when lending to a small business or an entrepreneurs was the same as when lending to the government… then we could talk about market rates. As is, to the cost of government debt, we need to add all the opportunity cost of all bank lending that does not occur because of the subsidy… and those could be immense.

I’m not going to get in that discussion here, but the point seems well-taken. But leaving that aside, I want to ask the following question: As long as the interest rate on TIPS is negative, why is the US government not selling massive amount of TIPS, using the proceeds to retire conventional Treasuries while earning a profit on each exchange of a TIPS for a Treasury?  That would be true arbitrage whatever the merits of Tea With FT’s argument.  Are there statutory limits on the amount of TIPS that can be sold?

The issue also seems to bear on the discussion that Steve Williamson and Miles Kimball have been having (here, here, and here, and also see Noah Smith’s take) about whether the Modigliani-Miller theorem applies to the Fed’s balance sheet. If there are arbitrage profits available exchanging conventional Treasuries for TIPS, what does that say about whether the Modigliani-Miller theorem holds for the Fed?

My next question is: if there are arbitrage profits to be made from such an exchange of assets, what is the mechanism by which the arbitrage profits would be eliminated? Why would exchanging Treasuries for TIPS alter real interest rates or inflation expectations in such a way as to eliminate the discrepancy in yields? Maybe there is something obvious going on that I’m not getting. What is it?  And if the reason is not obvious, I’ld like to know it, too.

UPDATE:  Thanks to Cantillonblog and Foosion for explaining the obvious to me.  In my haste, I wasn’t thinking clearly.  Given the expectation of inflation, the negative yield on TIPS will have to be supplemented by a further payment to compensate for the loss of principle due to inflation, so the cash flows associated with either a conventional Treasury or a TIPS are equal if inflation matches the implicit expectation of inflation corresponding to the TIPS spread.  But suppose the Treasury did issue more TIPS relative to conventional Treasuries, wouldn’t the additional Treasuries be sold to people who had slightly higher expectations of inflation than those who were already holding them?  Or alternatively, wouldn’t the very fact that the government was trying to sell more TIPS and fewer conventional Treasuries cause the public to revise their expectations of inflation upwards?  That’s not exactly the conventional channel by which either monetary policy or fiscal policy affects inflation expectations, but it does suggest that the policy authorities have some traction in trying to affect inflation expectations.  In addition, since interest rates fell close to zero after the financial panic of 2008, inflation expectations have responded in the expected direction to changes in the stance of monetary policy, rising after the announcment of QE1 and QE2 and falling when they were terminated.

UPDATE 2:  I am posting too fast today.  If the Treasury increased the quantity of TIPS being offered, it would drive down the price of the TIPS, increasing the real inflation adjusted yield.  An increased real yield, at a given nominal rate, would imply a reduced break even TIPS spread, or reduced inflation expectations.  Thus, increasing the proportion of TIPS relative to conventional Treasuries would induce savers with relatively lower inflation expectations than those previously holding them to begin holding them as well.  Alternatively, increasing the proportion of TIPS outstanding would encourage individuals to revise their expectations of inflation downward because the Treasury would be increasing its exposure to inflation.  But the point about the applicability of the MM theorem still applies with the appropriate adjustments.  At least until further notice.

27 Responses to “Why Not Arbitrage TIPS and Treasuries?”


  1. 1 Frank Restly June 4, 2012 at 3:36 pm

    “Given such low borrowing costs, Summers argues, governments should be borrowing like crazy to finance any investment that promises even a marginally positive real return, even apart from Keynesian stimulative effects.”

    Maybe you forgot what Larry Summers also said:

    “The federal government is a lousy venture capitalist”

    The role of a venture capitalist is to finance investment that promises a positive real rate of return (no investor likes to lose money to inflation – not even equity holders – see DJIA 1970 to 1980). If the federal government is lousy at identifying those investments and even worse at committing to those investments for a protracted period of time then Larry has some explaining to do.

    “As long as the interest rate on TIPS is negative, why is the US government not selling massive amount of TIPS, using the proceeds to retire conventional Treasuries while earning a profit on each exchange of a TIPS for a Treasury?”

    1. When exactly did the federal government become a for profit enterprise?

    2. Better question – why should the federal government sell bonds (nominal or inflation adjusted) at all?

  2. 2 Frank Restly June 4, 2012 at 4:20 pm

    “Given such low borrowing costs, Summers argues, governments should be borrowing like crazy to finance any investment that promises even a marginally positive real return, even apart from Keynesian stimulative effects.”

    “Consider a $1 project that yielded even a permanent 4 cents a year in real terms increment to GDP by expanding the economy’s capacity or its ability to innovate.”

    “As long as the interest rate on TIPS is negative, why is the US government not selling massive amount of TIPS”

    So Larry Summers wants the federal government to borrow more more money to finance investments that increase real GDP 4% a year, and you want the federal government to finance those investments by selling TIP’s that yield a negative rate of interest. Assuming that the negative rate of interest is 1% and nominal potential growth is 5% a year that leaves 1% inflation. Why exactly would anyone buy TIP’s that have a 0% rate of return after inflation – by design?

  3. 3 Cantillon Blog June 4, 2012 at 5:00 pm

    David,

    Exchanging issued nominals for issued TIPs is like selling breakeven inflation – compared to your starting position,you will be better off only if inflation is lower than what is priced in to the market at the time you do the trade(s). There are some technical aspects, but they are not important in order to understand the essence of the trade.

    US implied breakeven inflation at the 30 year maturity closed at 2.2026%. Last print for CPI was at 2.3%, which is not very far off where it has averaged the last twenty years. In the near term, further weakness in commodity prices (including for gasoline) has the potential to depress headline CPI further (and the lags mean recent crude weakness will start to come through shortly). However in the longer run, I suggest that the risks to inflation are very much to the upside in coming years.

    Still – the outlook for inflation is a matter of judgement. But the nature of the trade is a matter of analysis. Because inflation breakevens are not massively mispriced, present low real yields in TIPS are also reflected in nominals. Therefore if you want to lock in present depressed low real yields, you ought to undo recent policy of the Treasury and Fed (which has tended to significantly shorten the average maturity of the outstanding debt) and term it out, buying back 2 year notes and issuing more 30 years, and possibly even century and double century bonds. Whether you make these nominal or real bonds is a different question, but the point is your observations imply a duration shift, not a shift between nominal and real bonds.

  4. 4 foosion June 4, 2012 at 5:02 pm

    “As long as the interest rate on TIPS is negative, why is the US government not selling massive amount of TIPS, using the proceeds to retire conventional Treasuries while earning a profit on each exchange of a TIPS for a Treasury?”

    If market forecasts of inflation are accurate, the real returns on TIPS and treasuries are essentially the same (there may be an inflation risk premium on the treasuries and a liquidity discount on the TIPS, but these are likely small). TIPS + inflation = treasuries. In other words, there is no profit opportunity.

  5. 5 foosion June 4, 2012 at 5:07 pm

    ” Why exactly would anyone buy TIP’s that have a 0% rate of return after inflation – by design?”

    Because TIPS are safe and safety is very important in the current environment. If the government were to announce massive new programs that are likely to help the economy, then the optimism may raise the cost of TIPS. Where we reach equilibrium is not known ex ante.

    I see Cantillon Blog had a similar thought on arbing TIPS v nominals while I was typing.

  6. 6 cantillonblog June 4, 2012 at 5:08 pm

    foosion,

    If one is confident that one can earn a positive real return, then there is indeed a profit opportunity to term out one’s debt. But to do so one must buy back two year notes, and issue century bonds.

    I am not so sure that government finds it quite so easy to earn positive real returns on its spending. However I do believe that prudent debt management would indeed involve terming out of debt in nominal space. In fact, why not issue perpetuals such as the UK’s undated War Loan. In the battle to create inflation between the forces of deleveraging, and the central bankers, I am quite certain that the central bankers will win in the end.

    Of course terming out the debt is the opposite of the policy being followed in recent years by the Fed and the Old Lady.

  7. 7 cantillonblog June 4, 2012 at 5:14 pm

    Regarding, why somebody would buy a bond yielding a zero real return, that is a question I asked myself as a schoolboy on hearing about bank deposit rates in Britain in the 1970s. Why would anyone keep their money in an instrument that guarantees a small loss? Because at least it is liquid, and it may seem safer than the prospect of a larger loss. No force in nature guarantees that realized short term real interest rates over the life of the bond will necessarily be positive, and there have been many times when this has not been the case. Investing in a guaranteed negative real rate for two years may turn out better than taking your chances month by month (you might have even a more negative rate that way). Of course the alternative is to go out and buy a real asset, and hope it functions okay as a store of value. This is one of the evils of inflation and easy money – it either impoverishes the prudent, or forces them to speculate. And then when the bust comes, we heap scorn on the bankers for being reckless, forgetting the reason for their recklessness in the first place.

  8. 8 cantillonblog June 4, 2012 at 5:22 pm

    BTW one of the drivers of the bond rally in Europe has been that the solvency of pension funds and life assurance companies is assessed based on long-term swap rates or corporate yields. So the more that high quality debt rallies, the less solvent they become, and the more fixed income they need to buy and the more equities they need to sell and, as the market gets wind of these trades before they happen, this quickly becomes a vicious feedback loop. Pensioners are being forced to buy long-term bonds at economically insane levels based on regulations ostensibly designed to protect them. This makes no sense, practically speaking, but by the time the regulators and politicians have figured this out, the move will have burnt itself out. If we are lucky, at not too great a cost to future benefits.

  9. 9 foosion June 4, 2012 at 5:26 pm

    cantillonblog,

    Extending maturities when rates are very low does seem sensible. This should be the case even if markets are efficiently pricing the longer term debt or you believe in the expectations story of bond pricing.

    In an uncertain if not scary world, safety and liquidity can be very important. I’d rather be guaranteed a -1% real return than risk a more massive loss in some other investment, including a loss to inflation in nominal treasuries.

  10. 10 cantillonblog June 4, 2012 at 5:38 pm

    There is a difference between buying T Bills at a negative -0.20% yield for safety and liquidity, and buying 30 year German bunds at 1.70% because if you don’t you might go bankrupt on a mark-to-market basis and in terms of regulatory capital, even though it is not a question of cashflow. I suggest that trustees are not acting in the true interests of pensioners when they are forced by regulation and consultants of dubious competence to do the latter.

  11. 11 cantillonblog June 4, 2012 at 5:46 pm

    In the real world – I don’t know about the world of economists’ models – society is driven by emotional factors. A year ago people were worried mostly about inflation in Europe, and the ECB hiking rates. Today they are worried the whole thing will disintegrate, and people think the ECB will shortly cut rates again.

    People tend to say that we should replace the caprice of markets by the firm hand of government. The problem is that democratic governments are not better than markets – they are worse: an old lesson, of which we have been again reminded by recent developments in Europe.

    Still, one ought not to lose sight of the possibility that real prudence in government would in fact to be exerting a stabilizing influence. To term out the debt in the 1950s, when the English-speaking world seemed a safe place, free from the ravages of war and inflation that had befallen our less fortunate neighbours, would have been the height of sound debt management policy. And in the early 1980s, as fears of an inflationary apocalypse were pretty much mainstream, sound debt management would have involved shortening the maturity of debt.

    In theory this kind of approach is far from rocket-science. But one cannot in practice, and for political reasons conduct such a contrarian debt management policy within the confines of mass democracy. It is too subject to criticism from the gallery by those who have a greater voice than they do genuine expertise.

  12. 12 Woj June 4, 2012 at 5:55 pm

    cantillonblog makes some really good arguments on the matter. A further extension of the initial thought is what happens to interest payments if realized inflation turns out higher than the current break-even. If the US sells large sums of TIPS and inflation rises to even 5%, interest payments would quickly become a significant portion of current spending. Larger deficits would only push inflation higher, so government spending on other items would have to decrease significantly (or taxes could be raised). Therefore selling TIPS may as well be an indefinite commitment to low inflation.

  13. 13 dwb June 4, 2012 at 6:02 pm

    “why is the US government not selling massive amount of TIPS, using the proceeds to retire conventional Treasuries while earning a profit on each exchange of a TIPS for a Treasury ”

    isn’t that more or less just the portfolio balance channel for QE in the first place – replace some illiquid assets with liquid ones using the Fed balance sheet? normally its long term treasuries and cash, but this seems mostly a generalization (if you believe the “arbitrage” argument”).

    Maybe they don’t need to sell the TIPS, just buy massive amounts of treasuries and retire them (with currency) until inflation expectations were aligned?

  14. 14 Woj June 4, 2012 at 6:42 pm

    Selling TIPS also places the government budget at risk if inflation is significantly higher in the future. Interest payments would become a major portion of the budget and possibly constrain other spending. On second thought, that could be beneficial given the interest payments are largely just transfers.

  15. 15 Frank Restly June 4, 2012 at 8:07 pm

    “Therefore if you want to lock in present depressed low real yields, you ought to undo recent policy of the Treasury and Fed (which has tended to significantly shorten the average maturity of the outstanding debt) and term it out, buying back 2 year notes and issuing more 30 years, and possibly even century and double century bonds.”

    Very, very good…. Someone that understands a little bit about public finance. Because federal debt is accrual type rather than constant payment type, you are playing a game to determine if the market correctly prices in average maturity.

    Start with 15 trillion in debt with an average maturity of 4 years.
    Convert 1 trillion of 2 year notes to 100 year bonds.

    New average maturity after 2 year notes are retired = (15 trillion * 4 years – 1 trillion * 2 years) / 14 trillion = 4.143 years

    New average maturity after sale of 100 year bonds = (14 trillion * 4.143 years + 1 trillion * 100 years) / 15 trillion = 10.4 years.

    Done over the course of say 4 years this would be a 25.87% annual increase in the average duration of the U. S. debt.

    The present value for all taxes collected:

    PV = P x ∑ [ ( 1 + g ) / ( ( 1 + r ) x ( 1 + dD/dt ) x (1 – dAD / dt ) ] ^ i

    g = Growth Rate
    r = Average Interest Rate
    dD/dt = Change in rate of debt growth
    dAD/dt = Change in rate of duration growth

    Extending the average duration 25.87% per year out to infinity we get:

    PV = P x [ (1 + g) / ((1 + r) x (1 + dD/dt) x (1 – dAD/dt)) ] / [ 1 – (1 + g) / ((1 + r) x (1 + dD/dt) x (1 + dAD/dt)) ]

    For present value (PV) to equal infinity:
    (1 + r) x (1 + dD/dt) x (1 – dAD/dt) <= (1 + g)

    r <= (1 + g) / ((1 + dD/dt) x (1 – dAD/dt)) – 1

    Even with a 0% growth rate and debt increasing 7% a year the average interest rate on all debt would need to be:

    r <= (1 + 0) / ((1 + .07) x (1 – .2587)) – 1
    r <= (1 / .7931) – 1 = 26.07%

    And the federal government's revenues are enough to cover the interest payments.

    The break even interest rate for 100 year bonds (assuming a 0% growth and a non-infinite cost of money at infinity) would be:

    r < 26.07% * (100 – 1) / 100 = 25.81%

    The way you get there is by letting r(t) = Max Interest Rate * (t-1) / t
    Summing all interest rates to infinity and averaging them you get

    Average Interest Rate = Max Interest Rate * (1 – ( ∑ (1/n) / t ) ) where n = 1 to t and t approaches infinity. Taking the limit your Average Interest Rate is equal to your Max Interest Rate.

    But then again the cost of money at infinity is infinite since interest rates tend to follow a log curve. Adjusted for today's rates and following a log curve, 100 and 200 year bonds should yield about:

    2.5% * ln (100) / ln (30) = 3.39%.
    2.5% * ln (200) / ln (30) = 3.89%

    You would have to go out to quadrillion year bonds before you start seeing interest rates in the 25% range based upon today's sentiments.

    t = exp ( ln(30) * 25% / 2.5% ) = 590 trillion years.

  16. 16 Frank Restly June 4, 2012 at 8:35 pm

    Foosion,

    Maybe you missed what I was hinting at:

    ”Why exactly would anyone buy TIP’s that have a 0% rate of return after inflation – by design?” Because TIPS are safe and safety is very important in the current environment.

    Why would you buy TIP’s knowing that the federal government is going to try to use that money to increase real growth. Meaning the more successful their investment is, the worse off you are as an investor. Suppose instead of getting 4% real growth through its investment of your money, the federal government gets 6% or 7% real growth.

    There are lots of ways to just give your money to the federal government, for instance:

    http://www.treasurydirect.gov/govt/resources/faq/faq_publicdebt.htm#DebtFinance

    No sense inventing some cockamamy new methods.

  17. 17 Cantillon Blog June 5, 2012 at 1:45 am

    Restly – buyers of the long end at these yields are not buyers based on a speculative investment motive. They have to buy to match their liability benchmark, and don’t have the luxury of refusing to do so. Blame actuaries, consultants and regulators, as well as low quality, herding trustees.

    From a speculative point of view, I see no problem buying tips and hedging them with nominals at current pricing if you expect the putative plan to reflate to succeed. It’s the nominal long end that is most vulnerable – not tips.

  18. 18 David Pearson June 5, 2012 at 8:50 am

    David,
    I suspect you will be fond of this chart. It shows a rather high correlation between inflation expectations and S&P500 levels. As you might guess, the recent plunge in expectations has yet to be met by a similar decline in equities. If you are a believer in the influence of monetary policy on stock prices, and you think there are exceptions to EMH, would you take this opportunity to short stocks?

    http://soberlook.com/2012/06/equities-vs-inflation-expectations.html

    (Perhaps its a bit early to put on the trade — judging from the chart, the tail risk of a sharp S&P500 decline seems to increase significantly as inflation expectations drop to 50bp based on the 2yr/2yr break even — from 80bp today.)

  19. 19 Cantillon Blog June 5, 2012 at 9:00 am

    David Pearson,

    An interesting point, and I do not disagree with your conclusion. However one ought to distinguish between the two questions of correlation on a daily basis between breakeven inflation (not identical with expectations) and equity prices; and the longer-term relationship that will be realized over thirty years. There are of course some scenarios where realized inflation is high and stocks do very well, and some scenarios where it is low and stocks do very well (and similarly for bad case scenarios for equities).

    Governments ought not to try and trade, but to act with an eye on the long-horizon future. This is a unique niche not available to those who must act as agents to manage somebody else’s money. The Victorians certainly understood this, but I am not sure that moderns do!

    One further point – obviously some key factors for the inflation outlook are: crude prices, food prices, and wage setting behaviour. There is a strong short-term relationship between short-dated breakevens and energy prices. But study of history shows that food inflation is both necessary and sufficient for a persistent inflationary episode to be kicked off. We remember the 1970s for the oil shock. But that actually followed the hike in the price of soybeans, and I believe that it was this food price shock that started things.

    I suggest that we are early in a long-term bull market for food prices, and that it therefore makes sense for governments to structure their balance sheet so as to be long inflation. Buying back nominals and issuing same-duration TIPS is hardly prudent given this context.

  20. 20 David Glasner June 5, 2012 at 9:58 am

    Frank, The statement of mine that you quote was, like everything else in the post, written too hastily and not carefully thought through. My fault. However, what I was trying to say, and I think what Summers clearly meant as well, was not that the government should go out looking for private investment projects to fund, but to find public projects that promise even minimal net social benefits when discounted at a (near) zero rate of return. The reason that private investors aren’t undertaking these investments is that the returns generated by the investments aren’t appropriable by private investors, only by the government by way of increased tax revenues or fees. The point I was getting at is if there is an arbitrage opportunity because assets are mispriced, then the process of profiting from the opportunity should eliminate the opportunity. In equilibrium, the mispricing of assets is eliminated.

    cantillonblog, Thanks for the clarification. You are right that inflation expectations have changed mostly in the short term. For long time horizons, people assume that there will be some form of mean reversion, I guess, at least on the inflation side. Real and nominal rates have come down more sharply than inflation expectations over long time horizons. So, yes, it would seem appropriate to lengthen the maturity duration of the government’s debt. Or is there an information asymmetry? If the government (monetary authorities) are secretly planning to deliver less inflation than they are supposedly aiming for (forecasting), then wouldn’t they want to shorten the average duration of their debt?

    foosion, Thanks for your clarification as well.

    cantillonblog, Governments may not be able to appropriate all the gains from its spending in the form of increased tax revenue or fee revenue, but would we want our government not to undertake an investment that yielded positive net benefit to society just because the size of the deficit increased? Or phrased alternatively, if there are net benefits to society, there is no reason why a government could not extract part of those benefits in the form of increased tax or fee revenue. On pension funds and life insurance policies, even before Scott Sumner began lobbying for NGDP targeting, I favored indexing long-term government obligations, like social security payments to NGDP rather than to the CPI. Pensions and entitlement payments should not be fully protected in real terms. If real GDP goes down, pensions should also be adjusted accordingly. Why should some people be given absolute protection against risk by increasing the risk borne by everyone else? The point should be to share risk broadly not to eliminate it entirely for some people.

    Woj, The point is that to sell more TIPS you have to persuade people that inflation will be lower than they now expect it to be. Alternatively, by selling more TIPS, you are signaling that you intend to reduce inflation, thereby reducing expected inflation.

    dwb, The argument was wrong because there is no arbitrage opportunity on the exchange of TIPS and Treasuries as I was mistakenly thinking when I wrote the statement that you quoted. Now there could be an information asymmetry if the government knows that it is planning to deliver a different rate of inflation than it is ostensibly targeting, But if it tries to earn a profit from the information asymmetry it risks tipping its hand, so that the ability to profit from an information asymmetry seems limited.

    Will try to catch up with the rest of comments later.

  21. 21 cantillonblog June 5, 2012 at 10:37 am

    “I think what Summers clearly meant as well, was not that the government should go out looking for private investment projects to fund, but to find public projects that promise even minimal net social benefits when discounted at a (near) zero rate of return. The reason that private investors aren’t undertaking these investments is that the returns generated by the investments aren’t appropriable by private investors, only by the government by way of increased tax revenues or fees.”

    Yes – US infrastructure (bridges, tunnels, many airports, and the like) is rotting, and strikes many visitors as being more fitted to a third world country (where, funnily enough, these days everything is actually nice and new and shiny). There is no shortage of opportunities for a wise ruler to spend money on projects that would improve the commonweal. Alas, our American cousins made a decision some years ago to dispense with being ruled, and they are now at the mercy of democratic forces. If you look at where the stimulus money went, it is not clear that it was wholly channeled towards these kinds of long-dated projects. This being said, as the fashion for ‘austerity’ (the funny kind that doesn’t actually involve cutting overall spending) passes, I rather think we will see funding being channeled towards such projects, one way and another. Viable funding mechanisms remain somewhat obscure.

    “So, yes, it would seem appropriate to lengthen the maturity duration of the government’s debt.”

    Well – we now expose conflicts and questions of coordination between the debt management arm of the Treasury and monetary policy. Personally, I detest this accelerated debasement of the currency termed, for reasons of politeness, “quantitative easing”. (Credit easing is a distinct question). In my view, we would have been better off without it. But if you do support QE, you ought to consider the impact on the government balance sheet over time of selling low and buying high (which is what the governments of the US and UK have effectively done, at least with the government bond component of QE). Part of the reason yields are at this level is because the government bought back its debt at uneconomic levels, so if you supported the buying high it seems a bit funny to say “shouldn’t they actually being issuing new debt given it is so expensive”!

    “Or is there an information asymmetry? If the government (monetary authorities) are secretly planning to deliver less inflation than they are supposedly aiming for (forecasting), then wouldn’t they want to shorten the average duration of their debt?”

    There is not much of a constituency in the disinflationary camp, I would say!

    “cantillonblog, Governments may not be able to appropriate all the gains from its spending in the form of increased tax revenue or fee revenue, but would we want our government not to undertake an investment that yielded positive net benefit to society just because the size of the deficit increased?”

    Does the size of the deficit, and of future debt levels, not to mention non-binding entitlement promises have no importance at all? A certain number of people have been quite concerned about such issues of late – notably David Walker, the former Comptroller General – so it seems to me that you do not have the right to dismiss them without discussion.

    I shall not here bring up questions of public choice, of the efficiency of government programmes in delivering what it is intended they should produce (viz education and the Great Society), of the questionable nature of utilitarianism as a moral framework.

    “Pensions and entitlement payments should not be fully protected in real terms. If real GDP goes down, pensions should also be adjusted accordingly. Why should some people be given absolute protection against risk by increasing the risk borne by everyone else? The point should be to share risk broadly not to eliminate it entirely for some people.”

    Your argument strikes me as a bit funny in its emphasis. Since America’s independence, have there been any thirty year periods when real GDP has declined significantly? Yet the primary difference between indexing to nominal GDP and indexing to CPI is not the exceptional times when your favoured choice would tend to lead to benefits being lower than otherwise, but the more typical case when they would lead them to be more generous – given that real incomes tend to grow over time in spite of occasional setbacks.

    “dwb, The argument was wrong because there is no arbitrage opportunity on the exchange of TIPS and Treasuries as I was mistakenly thinking when I wrote the statement that you quoted. Now there could be an information asymmetry if the government knows that it is planning to deliver a different rate of inflation than it is ostensibly targeting, But if it tries to earn a profit from the information asymmetry it risks tipping its hand, so that the ability to profit from an information asymmetry seems limited.”

    Hmmm… Do we perhaps overestimate the knowledge of the government and its ability to achieve the inflation rate it desires? There is a fallacy known as the Texas Sharpshooter fallacy that applies here. With regards to putative information asymmetry, I am sure that you are familiar with the following (and this is one of many topsy turvy utterly mistaken comments uttered throughout the course of the buildup and unfolding of the crisis):-

  22. 22 cantillonblog June 5, 2012 at 10:40 am

    (I wanted to post Bernanke comments along similar lines, but picked the wrong link. Nonetheless, they are there if one digs a bit deeper).

  23. 23 David Glasner June 5, 2012 at 2:59 pm

    Frank, When real interest rates are negative, people buying TIPS are either forced by regulation to hold “safe” assets or are trying to the risk that inflation will erode the value of the conventional Treasury yielding whatever paltry return it now offers. Is holding a conventional Treasury yielding half a percent over 5 years less puzzling to you than holding a 5-year TIPS promising a negative yield?

    Cantillon Blog, That’s what”s so scary about the current environment, that people seem to be willing to pay such a high price for a 10-year or a 30-year Treasury, perceiving so little risk of taking a capital loss.

    David, Yes, it’s a lovely chart. Thanks. And I agree, the markets have been holding up rather well considering the fall in inflation expectations. I haven’t been running my regressions lately, so I don’t know how far off the recent data points are the regression line. I will have to check that and report back to you, but I have been awfully busy of late.

    Cantillon Blog, My view is that the positive correlation between inflation expectations and stock prices is atypical, possibly even pathological. Have you seen my paper and my various posts on the subject? You are right about the sequence of food price increases and oil price increases in the 1970s, but I don’t accept that they were both the result of monetary causes. The early 1970s involved a moderate 5-6 percent trend inflation with an oil supply shock, exacerbated by a dysfunctional program of wage and price controls.

    Under present circumstances, I am an outright inflationist. I don’t like quantitative easing, but there’s nothing else on offer. I am less concerned about the government’s balance sheet than I am actual GDP 10% or more below potential. If actual GDP were 10% higher the government’s balance sheet would take care of itself. You say that there is not much a constituency in the disinflationary camp, look at the inflation hawks on the FOMC and the all the warnings of imminent currency debasement emanating from prominent politicians.

    I don’t dismiss the size of the deficit and debt levels as matters of concern, but I think many people focus on it too narrowly and lose sight of the big picture.

    About indexing, I was just throwing out a thought that occurred to me as I was writing. Obviously a well-thought out indexation policy would not be pegged exactly to GDP. The idea would be to find some reasonable percentage of GDP growth that would provide some growth in benefits but would not give absolute protection against benefit reductions in times of economic contraction.

  24. 24 cantillonblog June 5, 2012 at 3:55 pm

    Yes, David, I have read your paper and various other remarks on the topic. I agree that the correlation we see today does not represent a healthy state of affairs. A financial academic might understandably attribute this to genuine fears over the possible although for now unlikely prospect of deflation ebbing and flowing on the basis of a noisy but fundamental sequence of financial and economic news. I don’t believe quite in this reductionist story, but it is better than many of the others, and you are right to draw attention to it.

    I believe that long-term cycles in growth and prices exist, and that we have basically passed through the point of liquidation of the Kondratiev winter in the private sector (the most intense part being the bankruptcy of the auto companies in 2008), with state government having been cut heavily too, and only federal government spending really lagging. If I am right, we should for a while see really pretty solid growth ahead in the developed and commodity-consuming world, although perhaps less so for commodity producers such as Australia, Brazil and Canada. In that scenario, I think equity market performance of related economies should continue to be somewhat resilient to the ebb and flow of price expectations. I think we are headed for the mother of all inflationary waves, but it will take quite some time to play out. Keep an eye on food prices in the coming decade or two.

    The problem, by the way, of dealing with a credit downturn when the banking system is very large in relation to GDP has been a little hairy in recent experience. But just imagine how much more difficult it becomes when nominal rates are higher and rising and we still have the same stock of debt in relation to GDP. The Fed can certainly buy Treasuries at auction to try and prop up the price. But it cannot do this and still retain control of the money supply.

    We were speaking of the constituency to realize inflation below the implied level in 30yr breakevens of two and change, and whether there were people that really wanted for their own peculiar reasons to deliver substantially less than this number. One verges into the territory of those who believe in green lizards at the heart of government here, because I am at a loss to know what imaginable reasons there might be to expect a secret plan for such (as you half-jokingly suggested). There are people who think the Fed has been too easy (and I am one of those in a certain sense, although it is too complicated to elaborate on here), but it is because they fear inflation will turn out much higher than is presently believed – not because they have a secret plan to push it much lower! Your scenario was discussing a deliberate plan to achieve a lower realised inflation rate in future. I don’t know anybody at the Fed who would like to achieve -2% starting from where we are now. If you believe otherwise, I should like to see your evidence!

    With regards to potential, does it not bother you that in practice it is really difficult to know what this is? My assessment of what went wrong in the 70s was that people were convinced potential output was much higher than, on reflection, and with the benefit of hindsight, it turned out to be, and as a result were putting coins in the fusebox to try and get there by hydraulic means. Inflation in both the US and UK (and actually I would say the same about the Eurozone) has been surprisingly resilient to those who believe in output gap theories of changes in inflation. Isn’t this a cause for concern given what happened last time around? One can point to some structural changes that mean we might have lost a fair chunk of capacity compared to what it seemed to be at the peak.

    And I really do question the sanity and good judgement of those who go on to estimate output losses by comparing output today, to what it would have been had it grown at the trend ending in 2007. Surely the whole point of a bubble is that you experience an artifical prosperity that people then take for granted, and plan around based on extrapolation of the recent past. In which case assessing potential output – as some have done – based on this measure leads one to a most foolish place.

    I think your point about sharing the pain of downturns, even for welfare beneficiaries, is a very sound one, even though it goes against fashionable notions about the value of fiscal automatic stabilizers.

    There is a more general problem of inference from the past in order to form an assessment of the future outlook. Studying economic history, whether or not one believes in the existence of cycles as such, it is clear that prosperous periods alternate with more difficult ones. Yet when one studies old newspapers, one is struck by how wrong people are – especially the experts – at the time of transition from prosperity to long-downturn and vice versa. It is human nature to extrapolate, but Nature’s nature slowly to mean-revert.

    This has clearly been the societal experience over the past decade. The boom led people to recklessness and overextension, and as a result we have to, or will have to make, cutbacks that we would not have preferred to had we properly understood the nature of growth.

    Perhaps well-conceived policy for the budget overall, and for welfare spending, would tend to lead spending to fall behind the general level in good times to build up a buffer, and to lag behind it in bad times to insulate us from Nature’s harshness. This might be commonsense, but it is not what currently happens.

  25. 25 David Glasner June 7, 2012 at 6:51 pm

    Cantillon blog, You are right that I was half-joking about a secret cabal at the Fed seeking to inflict deflationary pain on the rest of us, but I do believe that there is a certain central banking mentality that believes that it is costless or nearly costless to undershoot the inflation target and very costly to overshoot it, and I believe that view has been in the ascendancy since 2008, which is why the Fed tolerated a rapid contraction of the economy in third quarter of 2008 setting the stage for the financial crisis of the fourth quarter.

    Concerning the relationship between actual and potential GDP, I realize that there is some debate about that, but I think that the weight of opinion is still on the side that there is a big gap (10% is just a very rough estimate for discussion purposes on my part) between actual and potential GDP. I don’t believe that the loss of specific capital as a result of the downturn would have reduced potential output nearly as much as you are suggesting. If we had a substantial increase in the price level (in the absence of an obvious negative supply shock) without generating a significantly faster rate of growth in output, I would revise my view, but aside from doing such an experiment, I see no way to determine whether potential GDP was substantially reduced as a consequence of the 2008 downturn. If inflation went up rapidly without inducing a rapid increase in output and employment, one could then consider scaling back the monetary stimulus. One year of high inflation would not be disastrous. Even under the gold standard there were occasional bursts of inflation above 2 or 3 percent, without causing a loss of confidence and a wage-price spiral.

  26. 26 Cantillon Blog June 9, 2012 at 10:04 am

    Regarding the Fed’s inaction in Q3 of 2008, I doubt very much that they understood full well what was happening and schemed to err on the side of averting distant but not impossible inflationary scenarios. That wasn’t the case at all. Rather, as is consistent with their behaviour over the past dozen years, they had absolutely no clue what was about to hit us until it had arrived. One can soften this by pointing out that almost nobody else did either (at least amongst mainstream economists), but the point is they didn’t do something out of their normal behavior – they simply did not grasp how bad the situation was. This was a case of incompetence, not the malevolent and consistent pursuit of an institutional goal contrary to the commonweal. I say this having said at the time to people in the market that they wouldn’t cut enough until it was too late. The ECB was similar, always looking backwards.

    The weight of opinion was in 2005-2006 on the side of there being no housing bubble, or real problem in the credit markets. The IMF outlook for 2007 spoke of global synchronized growth, and a soft landing in the US. I remember being told that I should stop being so gloomy and go out and buy assets. GIven that the weight of opinion is consistently wrong at turning points, and we are debating whether we are in fact at one of these, I fail to see why the weight of opinion should be a factor supporting your case, and indeed one might hold this as something against it – and indeed I do.

    So, as I understand it, you have now refined your position from the original implicit claim that there was certainly a large output gap, to the more nuanced one that there probably is, and most people believe it to be the case, but we can’t be all that sure about it. In which case the risks of pressing down full speed ahead on the gas might be seen to be rather different considering the possibility that you (and the rest of the Establishment) might in fact be wrong. It seems we are back to questions of leads and lags in monetary policy. As I understand it you, or at least associates allied to your camp, have departed from the position that monetary policy has a lag of 6 to 18 months in taking effect. This once-consensus position implied one should be cautious about overegging the cake (and indeed provides some support for the ‘steady hand’ principle of the ECB, if not for its rather clumsy implementation). Whereas if you think there are no lags, then the costs of making a mistake is slim since you can correct it quickly without introducing oscillations.

    I do not mean this in an accusatory fashion, but do you find it surprising that inflation in the US, Europe, and UK has remained relatively high despite apparent excess capacity ? What do you make of it?


  1. 1 Lectura faina! | Dan Popa Trackback on June 7, 2012 at 6:16 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




About Me

David Glasner
Washington, DC

I am an economist in the Washington DC area. My research and writing has been mostly on monetary economics and policy and the history of economics. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey's unduly neglected contributions to the attention of a wider audience.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 442 other followers

Follow Uneasy Money on WordPress.com

%d bloggers like this: