Enter

Inquiry into Debt/GDP Feedbacks in the United States

In Economy on March 31, 2011 at 1:10 AM

What is the effect of United States government borrowing load applied to own-real income growth percentages? If we are to evaluate economists Carmen Reinhart, Maryland University, and Kenneth Rogoff, Harvard University, and their widely-cited, and now seemingly conventional, insight that debt levels breaching the implosive debt-to-gross domestic product limit of ninety percent literally imply significantly reduced real income growth for a country, then the United States audience should now be a captive audience in high-drama suspense mode. Worldly philosophers, withstanding, considered here-in is the application to the United States case.

Is the United States case a special case?

A major upfront distinction, and possibly self-preserving reward for America, is its fortunate dominance in happening to provide the world with a preferred world reserve currency, plenipotentiary-like in power. For example, by contrast, a country like Ireland, Greece or Iceland harbingers a large buildup of debt relative to GDP, progressively leveraged over time, only to quickly generate more damaging, negative feedbacks within the home economy, just about directly according to the Reinhart-Rogoff analysis, at least at the afore-mentioned 90 percent margin, i.e., GDP begins decreasing (and, synonymously, unemployment begins increasing) automatically right away. Relative to the United States, the narrative projects that these “smaller” countries effectively do not provide a dominant, preferred world exchange currency; therefore, they suffer disproportionately relative to the American case when implementing the exact same scale of debt load pursuits.

In particular, and importantly, the United States has just now breached the so-called Reinhart-Rogoff limit (see top panel in Figure 1 below). The question remains: Will the limit prove to rival Chandrasekhar’s astrophysical limit in intellectual power and legitimacy? But this time in the more immediate, visceral economic sense, cautioning us to avoid economic blow-up in advance of an ex ante predicted decline?

In balanced judgment, what caused the first global financial crisis?

In the wake of the Financial Crisis Inquiry Commission (FCIC) report that officially provides a Congressionally-sanctioned, neutrally-refereed explanation of the “avoidable” causes of the Epic Recession begun 2007-M12; apparently, now there are at least two required readings for Congress and policy makers in the States and the world over. The Financial Crisis Inquiry Report is now available in book version just as much so as Reinhart and Rogoff’s This Time Is Different, and/or associated working papers. Read together, in tandem, readers are provided with cause inquiries. In the former instance, a more qualitative, history-driven explanation is presented; in the latter instance, a more quantitative, economics-driven narrative is prepared.

President Richard Fisher, Federal Reserve Bank of Dallas, back in September 2008 stressed the crisis as being “the consequences of a sustained orgy of excess and reckless behavior” (Fisher, 2008). By contrast, others like Alan Greenspan, former Federal Reserve chairman, and William Black, University of Missouri – Kansas City professor and former litigation director of the Federal Home Loan Board, have placed causal burden in the hands of liars’ loans (Black, 2011) and control fraud (Greenspan, 2010). Former Securities and Exchange Commission (SEC) regulator Lee Pickard (2008) has pointed to the SEC’s removal of leveraging limits for the broker dealers as creating the impetus for generation of concentrated high-level lending losses.

Others too have blamed excessive borrowing relative to capital cushions. Fragility of the system has been criticized by Nassim Taleb, in particular. He wants Value-At-Risk models, i.e., VaR, originally popularized by JP Morgan, excluded from risk management practitionery (Nocera, 2009). The FCIC commissioners, i.e., Angelides, Borne and al, predominantly spotlight anti-regulation, however, both privately and publicly-organized, as the core malevolent antagonist.

Figure 1:

Interest in the Japanese Lost Decade:

Much has been written about Japan’s Lost Decade of the 1990s. The country represented, at the time of blow up, a comparably larger globalized economy and, hence, a meritable, worthwhile comparison point versus the current United States experience. However, the Japan case, nonetheless, breaks from American comparison at the same time because of: (1) the currency dominance particularity; and (2) the varyingly larger proportion of savings in Japan. According to Organisation for Economic Cooperation and Development (OECD) records, the Land of the Rising Sun brimmed over the Reinhart-Rogoff horizon in 1996 and has never managed to recede below ever since—in 2011 the estimate for Japan is to have brimmed over the 200 percentage point horizon by year end (OECD, 2011).

At some point diminishing marginal impact may kick in, but continuing declines in real income growth are proving otherwise, depending on how you measure your defense. After all, 2009 represented the largest single-year drop in real GDP growth on record judging from the OECD database (see Figure 2 below).

Figure 2:

Graphed above is a review of Japan’s economic Chandrashekr limit experience. As you can see, judging from the scatter plot of available time-series figures, as debt relative to GDP in Japan has progressed to surpass the 90 percent level, in sequitur terms, real income has largely managed to decay (with 2010 posing a most noticeable exception). Interestingly, and more rigorously, prior to 1996 the 10-year average for the latter variable measured 3.2 percentage points in real GDP growth versus 1.2 percentage points for the 10-year average afterward.

Continuing the comparison, below-produced in Table 1 are some simple, small-sample regression results methodologically akin to the results produced by Reinhart and Rogoff, presented in three separate models, discussing the Japanese real GDP growth percentages and gross debt-to-GDP level percentages in annual time-series terms, assessing the years from 1980 to present year.

Table 1:

Firstly, Model 1 presents a sample-size of 15 yearly observations comparing the relationship between the two variables when the debt-to-GDP level percentage is below 90. No proximate relationship exists judging from the results presented. Secondly, Model 2, by contrast, measures in simple linear regression the remaining 16 yearly observations where debt-to-GDP level percentages are greater than or equal to 90. Interestingly, this number of years outnumbers the lower level indebted number of years by 16:15. Similarly here, no relationship exists judging from the results presented.

Lastly, Model 3, evaluates the relationship presented in the table in a differently, by regressing real GDP growth percentages against a binary variable whereby 1 represents observations with debt-to-GDP percentages greater than or equal to 90 and 0 specifies observations with debt-to-GDP percentages less than 90. In summary, the binary results are highly statistically significant evaluated at the 99 percent confidence interval. Interpretation follows that annualized real GDP growth percentages are 2.69 percentage points lower when the 90 percentage point debt-to-GDP level is breached.

Additional analysis specific to the United States experience:

Figure 3:

Figure 3 provided above is a scatter plot representation of the underlying observations for the United States with real GDP growth percentages and gross debt-to-GDP percentages. Echoing the Japan evaluation, below-produced in Table 2 are simple regression results discussing the United States real GDP growth percentages and gross debt-to-GDP level percentages in annual time-series terms, assessing the years from 1940 to present year, a much longer time-period compared to the Japan analysis.

Table 2:

Firstly, Model 1 presents a sample-size of 63 yearly observations comparing the relationship between the two variables when the debt-to-GDP level percentage is below 90. No proximate relationship exists judging from the results presented. Secondly, Model 2, by contrast, measures in simple linear regression the remaining 8 yearly observations where debt-to-GDP level percentages are greater than or equal to 90. Differing from the earlier model, these results present statistical significance evaluated at the 95 percent confidence interval, that is to say, as debt-to-GDP increases by 1 percentage point, real GDP growth, in turn, declines by the ratio of 0.43 of 1 percentage point.

Lastly, Model 3, summarizes the relationship presented in the table in a different manner, by regressing real GDP growth percentages against a binary variable whereby 1 represents observations with debt-to-GDP percentages greater than or equal to 90 and 0 specifies observations with debt-to-GDP percentages less than 90. In summary, the binary results are marginally statistically significant evaluated at the 90 percent confidence interval. Interpretation follows that annualized real GDP growth percentages are on average 2.66 percentage points lower when the hypothesized 90 percentage point debt-to-GDP level is breached.

Academics and policy makers are periodically weighing in on the debate:

University of Georgia economics professor Jeffrey Dorfman surveyed the numbers on federal tax revenues and federal spending levels by comparing 2001 against 2010 (Dorfman, 2010). “We are currently spending $1.12 trillion more per year than what we were spending at the end of the Clinton administration, even after adjusting for population and inflation,” according to Mr. Dorfman. By contrast, “revenue is $420 billion lower today than it was in 2001 once we make the adjustments for inflation and population.” Professor Dorfman’s insight in mind, it is worthwhile to recall that former Massachusetts Institute of Technology (MIT) economics professor Robert Solow and U.S. Senator Patrick Moynihan alike, both Lords of Fiscal Finance, among others, reminded us that Congress has proven itself up to the challenge of improving the budget balance in the past, e.g., the 1990s period (Moynihan, Solow and al, 1995).

Noam Chomsky, a linguistics professor at MIT, has written on the so-called moral hazard dilemma uniquely afforded the United States by its preferential, WWII-inherited world reserve currency status (Chomsky, 2008). For her part, Mrs. Reinhart has not accommodated the possibility of American exceptionalism at play by offering distinction to the case of the United States in today’s context (Reinhart, 2010). As supportive evidence, of interesting note is that certain previous world currency originator failure episodes were indeed chronicled and incorporated within the debt/GDP study as cited by Reinhart and Rogoff, e.g., the United Kingdom in the pre-WWII period (Reinhart and Rogoff, 2009).

Mikhail Melnik, Georgia State University economics professor, opted for an arguably more artful portrayal of the opportunity cost stakes of American fiscal and monetary intervention via the United States government. “When there is a fire raging in the corner of the room, you do not take a priceless painting by Rembrandt and use it to put out the fire.” Apparently, he is of the opinion that the patterned emergence of financial sector-swelling within the overall income makeup of the United States has been a resoundingly positive event.

From other perspectives, Federal Reserve Chairman Ben Bernanke and Financial Crisis Inquiry Commission company appear more attuned to the argument for cause of GDP declines being routed in the development of the shadow banking sector (Bernanke, 2010); John Geanakoplos of Yale University places weight on the unwillingness of the Federal Reserve to regulate financial product downpayments, i.e., margin rates (Geanakoplos, 2010); Paul Krugman, Princeton University professor and New York Times columnist, is bewildered by reverse causality in the above-considered Reinhart-Rogoff estimations (Krugman, 2010); and David Lynch’s reporting suggests that world-reserve currency status appears to be more explanatory (Dionne, Jr., 2011).

Figure 4:

President James Bullard, Federal Reserve Bank of St. Louis, has written thoughtfully, granted in the context of advocating more asset purchases by the central bank–in effect, a policy that entails having the central bank absorb more government and private sector obligations as opposed to leaving the private economy to weather absorption of the borrowings (Bullard, 2010). More historically motivating, The Economic Consequences of the Peace (1919) by John Keynes, Cambridge University, makes for interesting reading right about now; especially, for someone advocating debt write-downs in the name of preventing a so-called “Carthaginian peace,” to quote the author, who walked out of the 1919 Paris Peace Treaty because of, in his view, punitive levels of reparations payments being demanded by French president Georges Clemenceau from the German government. Geopolitically, the obvious comparison would be modern-day China portraying 1919 France and the modern-day United States portraying 1919 Germany.

Elegantly put, leverage arguably offers the best proxy for core cause explanation. In the expert opinion provided by John Geanakoplos, he underscores a proven connection, for example, between average mortgage downpayment levels and average home price levels. He uncovers similar findings when analyzing leveraging central tendencies of securities dealers as well. Read his paper regarding managing the leverage cycle (Geanakoplos, 2010). Supportively, Federal Housing Finance Administration (FHFA)-provided data now presented on their public website, which includes information on GSE-acquired and private-label MBS-financed single-family mortgages with associated LTV characteristics, helps to reinforce such leverage cycle theory (FHFA, 2010).

Figure 5:

In the above-graphic, economer-calculated results are presented using similar methodology to that utilized by Geanakoplos (2010), whereby LTV information for the upper 50 percent lowest downpayment borrowers is averaged in order to generate the final results presented in leverage ratio terms. For example, in 2001 the average downpayment of 11.96 percent equates to a leverage ratio of 8.4 over 1 (phrased differently, this equals the 88.04 percent borrowed amount over the 11.96 percent downpayment amount). It is worth disclosing to the reader that these LTV estimates are more conservative than in Geanakoplos (2010) because underlying information regarding second mortgages, for example, is not provided by the FHFA. Piggyback borrowing automatically drives up overall leverage by increasing the overall debt burden. BankUnited, FSB apparently had borrowers that utilized simultaneous second mortgages (OIG, 2010).

FHFA addresses the understatement of overall leverage:

“The pattern of decreasing LTV ratios over time, most pronounced for loans financed with private-label MBS, is consistent with the greater use of second liens to avoid mortgage insurance on low-down payment mortgages, a practice that was increasingly common into 2007. In addition, loans with LTV ratios at origination of 80 percent or 90 percent tend to have higher delinquency rates than loans with slightly higher LTV ratios in several origination years. That observation is consistent with the existence of second liens that are not captured in the LTV ratio.”

Additionally, the sub-section of the FHFA data concerning the highest levels of LTV classifications is conservatively weighted in an overall index, adjusted to the minimum FHFA-provided value of 105 percent. Interestingly, in 2007 the home price index peaks in the simultaneous year of maximum average leverage. In particular, leverage topped out at 10.6 at this point. By that year, home prices had risen over 45 percent compared to 2001.

Correction: March 25, 2012

An earlier version of this article contained the misquote, “If you have a fire in the corner of the room, you don’t take the Rembrandt painting off the wall and throw it at the fire.” The misquote has since been corrected to reflect the accurate quote according to Mikhail Melnik, “When there is a fire raging in the corner of the room, you do not take a priceless painting by Rembrandt and use it to put out the fire.” The author expresses his sincere apologies to Mr. Melnik.

  1. First of all, I want to say that it is always a very positive development when someone becomes so passionate about economics. You present here a very interesting and timely discussion. My only point is that when you quote someone make sure it is the correct quote. As you might recall my statement was: “When there is a fire raging in the corner of the room, you do not take a priceless painting by Rembrandt and use it to put out the fire.” The statement was made to illustrate that certain permanent institutions, such as the market principles, should not be sacrificed in order to resolve although a very powerful, but short term crisis.

    As you might recall, I made three references to the Federal Reserve in my presentation. One, the housing bubble that burst in 2007 was “effectively” shifted by the Federal Reserve from the NASDAQ bubble during 2001 – 2005. The Federal Reserve used the interest rate policy to help reduce the effect of the NASDAQ bubble burst that began in March of 2000, and in the process caused the housing bubble to form. Two, the Federal Reserve failed to foresee the burst in the housing bubble. This is seen in the GDP data where you see housing investment starting to decline in January of 2006, while the monetary policy starts to be implemented nearly two years later and after the summer cycle of the housing market. Three, the recent monetary expansion round was aimed at lifting certain asset prices, in particular the housing prices, however, with inflation, one can only unleash it, but not control it. As a result, the monetary expansion helped create commodity inflation, even a stock market appreciation, but appears to have had a very limited effect on housing values.

    In any event, I enjoyed reading your post, and I am very happy to see that people do take economics so passionately; I only wish you had my citation right.

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

%d bloggers like this: