Lacy Hunt has used econometric research to persuasively demonstrate the statistical relationship between excessive debt and slow economic growth. Although Hunt and I disagree over whether this analysis can be applied to the U.S., our forecasts for growth in the U.S. economy and for the bond markets are remarkably similar.
Hunt spoke May 15 at the Strategic Investment Conference in San Diego, sponsored by Altegris and John Mauldin. I also spoke with him by phone May 22.
A copy of Hunt’s presentation is available here.
In prior presentations, Hunt has focused mostly on research, such as that by Carmen Reinhart and Ken Rogoff, showing the relationship between public-sector debt and slow growth. In this presentation, he reviewed studies that showed how high public- and private-sector debt also damp growth.
Not only does growth slow, Hunt said, but once a country’s public- and private-sector debt exceeds 275% of GDP, the relationship becomes non-linear, with growth slowing at an ever-increasing rate.
The questions are whether this analysis can be extended to the U.S. and, if so, what implications it carries. I’ll review Hunt’s findings – including what I believe are its limitations – and where both he and I think the U.S. economy and bond markets are heading.
From 90% to 275%
Until April of last year, most of the discussion on the relationship between debt and growth was based on Reinhart and Rogoff’s 2010 paper, Growth in a Time of Debt, which argued that a country’s median growth rate falls by 1% once its debt surpasses 90% of its GDP. That was until a University of Massachusetts graduate student, in a highly publicized episode, found errors in the spreadsheet used for the study.
Hunt said that Reinhart and Rogoff’s results are still valid, in part because their findings were based on median results that were unaffected by the UMass corrections. He said other research using a larger group of countries, published in 2012, showed an even stronger relationship between debt and growth. Indeed, Hunt said, there are another 15 studies validating the Reinhart and Rogoff findings.
His talk in San Diego drew on a 2010 study, The Impact of High and Growing Government Debt on Economic Growth: An Empirical Investigation for the Euro Area, by Cristina Checherita-Westphal and Philipp Rother, which shows a 275% threshold for public- and private-sector debt that triggers slow growth. Hunt said this study is significant for two reasons: It demonstrates the non-linear effect above 275% and it makes a strong case for causality.
Despite their power, econometric studies cannot prove causality. Hunt and I agree on this point, and we agree that statistical relationships found in regression analyses are necessary to establish causality. This study made its claim for causality by “identifying the channels through which the level and change of government debt is found to have an impact on economic growth: (1) private saving, (2) public investment, (3) total factor productivity and (4) sovereign long-term nominal and real interest rates.”
Checherita and Rother found statistical relationships between those four factors and growth rates. Since each is closely tied to the ability of a country to service its debt (e.g., higher interest rates increase the debt burden), they were able to argue that high debt causes slow growth, not that the two were merely statistically correlated.
When debt starts to hurt
While those studies, along with others Hunt cited in presentation, make a strong argument for causality, they don’t prove that their results can be applied outside the data sample studied.
Checherita and Rother studied 12 European countries since 1970. Most belonged to the European Union and could not control their own currencies. None of those currencies was the reserve currency, a status only the U.S. dollar enjoys. Similarly, Reinhart and Rogoff mixed small and big countries.
I believe the U.S. is a special case – by definition, a statistical outlier – to which these types of studies cannot be applied. Because it controls its currency and can and will be able to borrow at very low real rates, there is no reason to expect that the U.S. will face a non-linear deceleration in its economic growth because it has surpassed 275% of public- and private-sector debt-to-GDP.
The U.S. reached that level in 2008 and is now at approximately 350%, having declined slightly in the last year.
Hunt acknowledged one problem that Checherita and Rother’s study shares with all other econometric analyses of the relationship between debt and growth: They don’t look at how the money raised by issuing debt was spent. If that money is spent productively – on projects that exceed the cost of capital – they will stimulate growth. Money spent unproductively – for example, to service existing debt – will have the opposite effect.
The economist Hyman Minsky is known for his contributions on this issue. Minsky distinguished between hedge finance, speculative finance and Ponzi finance. Hedge finance, Hunt said, generates an income stream that can repay principal and interest. Speculative finance can repay only interest and Ponzi finance repays neither one.
The U.S. currently spends about 6% of its revenues on interest payments on its debt. I asked Hunt whether that level of spending was sufficiently unproductive to fall into the category of speculative finance that would damp our growth. He said in the U.S., Europe and Japan, promises are being made to finance non-discretionary obligations that will later prove impossible to maintain. “The total non-discretionary pie is crowding out the good stuff,” he said.
“There is a level at which the debt becomes counterproductive,” he said, “At low levels, you would presume that you're skewed enough toward the productive debt, and when the debt goes higher, you're skewing it more and more to the unproductive or counterproductive debt.”
Nobody would disagree with this, but reasonable people can disagree as to whether the U.S. has reached a level that makes its debt counterproductive.
What future does the U.S. face?
Harsh critics of U.S. debt levels typically claim that it will face a “Minsky moment” – a point where it will be forced to default on its debt, its interest rates and inflation will surge or its currency will be devalued. I asked Hunt whether he foresees such a cataclysm for the U.S.
He predicted another path: the Japanese outcome.
Those same deficit critics invariably have a hard time explaining what has happened to Japan over the last three decades. Its debt-to-GDP ratio has soared, yet its Minsky moment hasn’t come. It has experienced painfully slow growth but no collapse in its bond market or currency. (Hunt said that Japan will someday face a Minsky moment, but he didn’t say when.)
Japan is the cautionary tale for the U.S., according to Hunt, as it is for Europe. He foresees an economy with lower private savings, less investment, decreasing productivity and an erosion in standard of living.
Overall, I agree with Hunt on this point, but for different reasons. Because of its reserve-currency status, the dollar will remain strong relative to other currencies. This will cause the U.S. to run a persistent trade deficit, with jobs being lost to lower-cost competitors in other countries. Unemployment will remain high and domestic growth slow. Low inflation – or possibly deflation – and low interest rates will follow. I disagree that the U.S. will face a decrease in its standard of living, though, because our strong currency will allow us to purchase cheap imported goods.
Hunt and I also agree on the best course of action to stimulate growth. Hunt cited Thomas Hobbes, the pre-Renaissance philosopher, who said that income measures what you contribute to society, and spending measures what you take from society.
“What we really need to do is incentivize income and dis-incentivize consumption,” he said. “We need a period of stronger income growth in which spending lags, so that we can de-lever the aggregate balance sheet.”
The most direct way of accomplishing this would be through a value-added tax (VAT) that is appropriately progressive. The VAT would need to be counterbalanced by a decrease in income-tax rates.
Hunt calls this austerity – a multi-year increase in the saving rate – which is the definition put forth by the McKinsey Global Institute. But that is not the same type of austerity that deficit critics call for when they ask for reductions in government spending – policies that have had mixed results in Europe.
As a fixed-income manager, Hunt has positioned his portfolio consistently with his economic forecast. His portfolio has a duration of 20, which is a big bet on lower long-term interest rates. He said he would turn bearish on bonds only if U.S. debt levels moved significantly toward or below 275% of GDP.
He considers that unlikely. Barring a major technological revolution – something on the scale of replacing the internal-combustion engine – he predicts a parallel to Japan’s lost decades, a forecast with which I unfortunately agree.
Read more articles by Robert Huebscher
