According to former Secretary of Treasury Paul O’Neill, Dick Cheney is reputed to have said: “…deficits don’t matter.”
(see Suskind’s The Price of Loyalty, and online here). What’s the (updated) evidence?
Updating an earlier paper (discussed here), Jeffrey Frankel and I gathered data on ten year Treasury yields over the 1988-2006 period, as well as the debt to GDP ratio expected two years hence, as projected
by the OECD. In order to control for other factors, We include the inflation rate over the previous year, as well as the output gap. We then estimate the following regression:
il = β0 + β1Ed + β2π + β3gap + u
Where:
- il is the ten year Treasury yield;
- Ed is the debt to GDP ratio projected two years ahead;
- π is the y/y CPI inflation rate;
- gap is the output gap in percentage points.
OLS yields an adjusted R-squared of 0.48, DW of 0.72. Using robust
(Newey-West) standard errors, one obtains a coefficient on Ed
of 0.076 (t-stat of 2.43) — each one percentage point increase in the debt-to-GDP ratio projected two
years hence induces a 0.08 increase in the long term yield. The inflation coefficient is
0.92 (t-stat of 4.05), and the output gap coefficient is is 0.26 (t-stat 1.00).
I conclude from this that debt matters. At the same time, I would not claim
that the estimates are terribly robust. One interesting result is that the prediction
errors are quite large and negative beginning in 2004 (the dashed lines are plus/minus one
standard error of regression residual).
Figure 1: Actual ten year benchmark bond yields, predicted values, and residuals (with plus/minus one standard error).
Source: OECD, Economic Outlook (various December issues) for debt-to-GDP ratios, output gaps, and IMF, IFS for interest and
inflation rates; and author’s calculations.
The gap might be due to the influence of capital inflows, as argued by Warnock and Warnock, as well as Frankel and myself (who focus on official inflows).
Interestingly, the prediction error is shrinking in 2006.
[Late Addition 10:20 am, per Brad Setser’s request]
Figure 2: Prediction error (blue) and gross purchases of Treasuries as share of GDP (red). 2006 data for first three quarters. Source: Chinn and Frankel (forthcoming), author’s calculations.
Technorati Tags: deficits, debt,
interest rates,
inflation rate, output gap.
It sounds like you have rediscovered a rule-of-thumb about long-end interest rates that has been around for some time. Major influences, in descending order, are inflation, growth and fiscal policy. Only by correcting for the first two, which have a larger influence, can you see the impact of fiscal policy. Gazing at a nominal yield chart is not a good way to correct for the first two influences, but it seems to be all that many politically motivated “analysts” have been willing (able?) to do.
Menzie — we should do a joint post super-imposing your error term with three variables:
BEA official inflows
My measure of $ reserve growth (which will probably have a good fit)
Foreign treasury purchases — official and non-official. (not as good a fit for 06 as they fell off .. with more demand going to agencies)
i suspect the correlation between your error term and $ reserve growth will be striking.
If the impact of an additional $130B in deficits (~1% GDP) is only 8 basis points on the long end, it seems like deficits don’t matter a whole lot. Am I interpreting your regression coefficients correctly?
Don’t demonstrate that deficits do not matter, it might keep the Democrats from using deficits to finance Medicare.
I sure that is why Cheney wanted to demonstrate that deficits do not matter, Right?
kharris: I will admit that our regression specification is not innovative, except to the extent that we use expected debt (which in turn follows Laubach, and others, as cited in the Chinn-Frankel paper).
Brad Setser: (Partly) Done. Didn’t have all the numbers at my fingertips, but did include the gross Treasury’s purchases/GDP ratio. This is an update of a graph in the Chinn-Frankel paper that is linked to in the post.
STS: If deficits exhibited no serial correlation, your interpetation would be largely correct. However, a permanent increase in the deficit induces a larger increase in the debt-to-GDP ratio.
anonymous: I have no idea what goes through the Vice President’s mind, when it comes to economics.
So deficits don’t matter a lot in the short run, but the crowding out effect in debt markets can have a lot of cumulative impact because of the serial correlation. That corresponds to my prior intuition about the question.
Conditioning on being a politician, this translates to “deficits don’t matter” since the conditioning truncates concern for the “long term” 😉
In the aggregate you could use statistics to “prove” that deficits do matter, but a much more important question is what created the deficit. Many households run deficits at some point in their lives especially when the family starts a business or purchases a home, but at such a time a household will reduce spending either voluntarily or by coercive action of creditors to ruduce or end the deficit.
In today’s world of fiat currencies the government need never run a deficit. The government can simply print money and never run a deficit. Granted this will cause inflation and the analysis is adjusted for inflation, but the inflation does not become fully manifest until months, even years after the injection. For this reason analysis of a government deficit is simply an academic exercise.
But consider a deficit caused by reducing the tax and regulatory burdens on production allowing the free market to increase production and income. The deficit will be temporary, much like a household investment in a business or a home, but tax receipts from increased business profits will quickly overwhelm the deficit. This would have happened in the 1980s had congress not increased spending at twice the rate of the revenue increase, and it was seen in the 1990s as the growth of spending was reduced especially by the reform of welfare and the decrease in capital gains taxes. Then once again in the 2000s as tax reductions increased business income and tax receipts after 9-11 we weathered a series of the worst natural and man-made disasters our nation has ever seen.
An economist superfically analyzing the deficit without digging into the underlying causes is like a doctor predicting a patient’s death by only reading his temperature.
DickF: If we don’t rely upon data, then all we are left with are anecdotes — and faith. Right now, I’m not certain you have a falsifiable hypothesis, given your skepticism regarding econometric analysis (although if you write out the model, we’ll be able to make a determination). So for me, I’ll stick with data.
KHarris has a good summary. I think everyone has to admit that deficits matter; but it’s not necessarily the case that they matter much.
The early 1980s are a really rich time period for this kind of study; I’d feel a lot better looking at coefficients estimated from late 1970s forward. I assume that OECD deficit predictions are the limiting factor on the data set. Any good alternatives?
I have no problem with someone making a living by playing SUDOKU.
Economics can be considered either a mathematical science or a behavioral science. I believe that economics is primarily about satisfying wants and needs and so is a behavioral science. I do not want to eat hamburgers for every meal. Because of this I do not believe that econometrics should be the primary focus of economics, that focus must be human action and decision making and how they are influenced by government. Econometrics is useful in economics when it is used to explain properly defined conditions and circumstances, but in a macro context it is often simply a mathematical game. Changes in policy change behavior dynamically but mathematical formulae remain static and in the macro miss the disasterous effect of extreme variations from an acceptable mean.
My purpose of posting here is to give an alternative vision of economics and to encourage deeper thinking beyond just the mathematical equation. Hopefully I will be successful and will stimulate discussion as I challenge the dogma of this site.
Let me state that I appreciate what you and James Hamilton do here. It is important to discuss these issues and to get the discussions into political decision making because these issues make a difference our lives, even the lives of non-economists, but it is important in academe to consider alternative views if we are to grow intellectually.
Menzie-
A couple of thoughts without reading the paper (let me know if the paper addresses – I apologize) … first, the prediction error does not look random … suggesting that the model fit is not that good … your comments that other factors are at play seem more than very likely … second, the model you forward has multiple time horizons … wouldn’t you expect correlations with these variables and error terms … I do intuitively … In summary, I not sure the model is that good … intuition and Keynesian reasoning suggest your conclusions … I do not think the interest rate time series suggests the relationship.
How much money the government spends matters. What the money is spent on matters. Government spending draws real resources away from the private sector, and affects the general economy.
How the government obtains its funds matters. The level of income taxes matter. The level of other taxes matters. The amount of borrowing by the government matters. All of these things determine who will forgo real consumption so the government can have those real resources. This affects the economy.
The deficit is a derivative of the above factors. Does the deficit matter? – I think so. But all of the components of the environment that result in a deficit are far more significant than the net result of the funding math.
The allocation of real resources is more important than the monetary measure of it.
The government has taken the real resources and someone has lost those resources – but a portion of that real transfer appears to be borrowed. However, if the debt is not repaid the real transfer is permanent.
Truthfully, I think a deficit is really a hidden tax which manifests in various ways. But we cannot reliably measure or prove its impact. Very few people even realize that they are paying it. It is the perfect tax.
bill conerly: The time sample is restricted by the availability of comparable two-year-ahead projections across countries from the OECD. An excellent, US-specific, analysis (now a little dated) is by T. Laubach, “New evidence on the interest rate effects of budget deficits and debt,” Finance and Economics Discussion Paper No. 2003-12 (Fed Board, March 2003).
sonya: The regressions up to 2004 we estimated in the cited paper (Table 4) have DW statistics of 1.52 to 2.12; the standard errors were Newey-West robust standard errors. Extending the data up to 2006 does yield an equation with more serial correlation, which supports the view that an additional variable is important in recent years (motivating the inclusion of Treasury purchases in the paper’s regressions).
DickF: Thank you for the compliments. I believe you mis-apprehend my approach to economics. I too believe it’s primarily a behavioral science, and indeed in the spectrum of economists, I think I am considered on the “non-mathematical” or “non-theoretical” end. I believe that it’s difficult to model empirically macroeconomic phenomena, but, to paraphrase Nixon, I also think we are all now Lucas-ians, to the extent that we understand the power of the macroeconometric policy critique. At the same time, we understand the limitations of that critique (see e.g., Rudebusch (2002)).
Finally, I would not characterize the material I post as dogma — my teachers have included Leibenstein (x-inefficiency) and Akerlof (near rationality; the market for lemons). I’ve written for sociology and policy journals. But I admit what I value are arguments that can be documented empirically. Otherwise, the discussion is susceptible to devolving into argument by assertion.
zephyr: I agree. But my limited intellectual capacities (and time) allow me to only address one issue at a time.
Deficit do matter. And there are WMD’s in Iraq; just not the one’s the media talks about.
WMDs FOUND IN IRAQ
Did you know that U.S. troops have fired in combat 2,500 tons of radioactive bullets and shells during the past 16 years? Made from a material referred to as Depleted Uranium (DU), these kinetic penetrators are 99.8 percent Uranium 238, a radioactive emitter of Alpha radiation with a half-life of 4.5 billion years. Montana Depleted Uranium Education Project
The United States government sent Dr. Doug Rokke to the Middle East during and after the first Gulf War. He was sent to investigate the effects and amounts of DU left there by our military. Upon returning home, Dr. Rokke became sick from the very substance he was sent to investigate.
We found the Weapons of Mass Destruction and Dirty Bombs in Iraq. They are ours! We have found the enemy and he is us. [more]
QuestionItNow Still In Iraq
Menzie,
Just so you understand my use of words I am using dogma in the above post to mean “something held as an established opinion.” Any group discussing issues has an established dogma that they honor subconsciously. There are certain accepted “truths” that are accepted a priori. What I hope to question are these a priori assumptions. I did not intend to attack you.
DickF: No offense taken, at all. But I think some readers have the view that both Jim and I are hopelessly mired in standard economic analysis, so it’s useful to emphasize that Jim and I don’t agree on everything, and I like to think my set of priors are not easily pigeonholed.
Interest rates are only one thing that is affected by debt. Also affected are inflation and GDP growth. Lowered GDP growth in turn increases the debt in a positive feedback manner. Now I realize that all of these are tied together, but small changes in interest rates are not the only thing to worry about.