A working paper by the Bank for International Settlements entitled "The real effects of debt" by Stephen Cecchetti, M. Mohanty and Fabrizio Zampoli, is a particularly pertinent bit of research in this time of troublesome levels of both personal and sovereign debt. In this paper, the authors seek to answer the question "When does debt go from good to bad?". The authors used an extensive dataset for OECD nations from the years between 1980 and 2010 to examine the levels of government, non-financial corporate and household debt and from that, derive a conclusion that gives us some idea of when debt levels reach the danger point.
The authors open by noting that "debt is a two-edged sword". Most of us have used debt at one time or another to purchase a car, house or other major consumer item, we are well aware that corporations issue bonds to fund capital expenditures and that governments seem to issue endless reams of debt to fund both capital and operational expenditures. As we all know, too much consumer debt is a bad thing, a lesson that many unfortunate consumers in the United States have learned the very hard way. We are now starting to see the tide turn against nation after nation around the world as their profligate spend and tax philosophies come home to roost, forcing nations to adopt very painful austerity measures. The mounting levels of sovereign debt have and will continue to force many nations to make painful program cuts in what have traditionally been regarded as essential services. Apparently, politicians, business owners/executives and households are learning that debt beyond a certain level really is bad for growth as we see in mainstream media headlines (perhaps a bit too late!).
Debt is used for good as a mechanism to smooth consumption, investment and taxes. In the case of individuals, assuming debt allows households to continue to purchase goods and services as their income levels change. In the case of corporations, assuming debt allows them to continue to make investments in equipment and production in the face of intermittent declines in sales. In the case of governments, assuming debt allows them to smooth taxes as revenue levels vary and can also assist governments in smoothing expenditures across generations. Unfortunately, assuming increasing levels of debt increases the risk that repayment may not take place, particularly in an environment where interest rates rise and fall. Rather than smoothing the economy, debt, when excessive, can result in economies that lurch between booms and busts driven by threat of default. This is what is being experienced by much of the Eurozone right now.
The graph on the left show the changes in the world's nominal level (not corrected for inflation) of household, non-financial corporate and government debt (termed non-financial sector debt) over the past 30 years as a percentage of GDP:
According to the authors, the level of all three types of debt has risen relentlessly over the past three decades, from 167 percent of GDP in 1980 to 314 percent of GDP in 2010, a rise of 5 percentage points per year over the 30 year period. Governments account for 49 percentage points of the increase, corporations for 42 percentage points and households for 56 percentage points.
Now look at the graph on the right where the debt data is corrected for inflation. Real government debt has risen by about 4.5 fold (for an average annual compounded growth rate of 5.1 percent), corporate debt has risen by 3 fold (average growth rate of 3.8 percent) and household debt has risen 6 fold (average annual growth rate of 6.2 percent) for an overall annual average compounded growth rate of 4.5 percent.
Looking more carefully at the graph, you will notice that the rise in household debt (brown line) outstrips the rise in both corporate and government debt. That's a tad concerning, isn't it?
Here is a chart that shows a country-by-country breakdown in total debt accumulation over the 30 year period:
Total debt in Japan in 2010 exceeded 450 percent of GDP, 350 percent in Belgium, Portugal and Spain, 260 percent in the United States and 313 percent in Canada with the median value at 322 percent.
Here is an interesting graph showing the breakdown between the growth in public and private debt in the United States since 1950:
Notice that the level of total private and public debt remained relatively steady at 150 percent of GDP (with the level of public debt as a percentage of GDP actually declining) for the first thirty years, the level of debt began to rise in the early 1980s, restabilized at about 200 percent of GDP until the early 2000's when it began to experience its most rapid and long term increase to nearly 270 percent of GDP.
Why has borrowing increased so much? First, an prolonged period of relatively strong economic growth and low inflation lulled consumers (in the United States in particular) into a false sense of security. Consumers consumed more and more, necessitating borrowing more from lenders only too willing to oblige. As well, a rather substantial decline in real interest rates made the "drug of borrowing" much easier to swallow. In addition, tax policies including the deduction of mortgage interest, encouraged consumers to borrow more as home ownership expanded, a tale that did not particularly end well.
Now, let's answer the question "when does debt reach dangerous levels"? At what level does high overall debt level begin to impact economic growth? Here is a graph that at least partly answers the question:
The authors took the mean per capita GDP growth levels (in light brown) for each of the 18 OECD nations and plotted them against the mean of the non-financial sector debt as a percentage of GDP (in khaki) which were divided into four quartiles or groupings based on size. Mean per capita GDP growth rises as debt rises until we reach the fourth quartile (highest debt-level nations with mean debt-to-GDP in excess of 300 percent) where GDP growth shrinks noticeably. This shows that as non-financial sector debt rises, output rises but only to a certain point when excessive debt results in shrinking output. Using more sophisticated regression analysis, the authors show that there are certain statistical debt thresholds beyond which the levels of government, corporate and household debt will impact economic output. For governments, the debt threshold is 84 percent of GDP, for corporate debt, the threshold is between 75 and 90 percent depending on whether controls are in place for banking crises and for household debt, the threshold is estimated at 84 percent. In the case of government debt, when the debt exceeds the 84 percent threshold, an additional 10 percentage point increase in debt-to-GDP will actually drive growth downward by 10 to 15 basis points (0.1 to 0.15 percent). This additional debt has the exact opposite impact on the economy that governments want.
Many governments around the world find themselves in the unenviable predicament of "owning" a debt-to-GDP ratio that is in excess of the aforementioned threshold values. Perhaps that explains why the stimulus programs of the past 3 years have really done very little to boost economic growth over even the short-term. This means that governments basically have no choice but to reduce deficits and future debt liabilities, an issue that is rarely, if ever, discussed. Unfortunately for all of us, this is highly unlikely since demographic changes are working against any austerity measures. Aging populations have the unfortunate impact of driving up expenditures just as revenues are falling, making a bad situation far, far worse. As shown on these last two graphs, dependency ratios (the non-working-age population as a percentage of the working-age population) are already rising in many "advanced" nations and are projected to rise until the mid-2050s:
As many of us are aware, this is the demographic ticking time bomb that is already having an impact on Japan, the holder of the world's second largest nominal debt. Japan saw its dependency ratio being to rise in the early 1990s due largely to one of the world's lowest fertility rates as shown on this graph:
Nations with populations that are not replacing themselves all face the same issue at some point in time. Europe and the United States are on the cusp of the dilemma now. Fortunately for the world's emerging economies in Asia and Africa etcetera, they have been given a demographic reprieve until the mid part of this century for the most part. Unfortunately for the rest of us, many projections show that unless fiscal policies change very soon, debt-to-GDP levels will explode as the demographic demons come home to roost and baby boomers line up to collect their entitlements.
High levels of government debt become dangerous when the ability of government to raise tax revenues to both service and repay the debt is questioned. We are at that point now. No one wants to pay more taxes only to see them frittered away by politicians with a finely honed sense of entitlement. The impact of excessive debt held by corporations and households complicates the issue further. The authors of this study conclude that their is a very clear link between too much debt and a lack of economic growth. With the demographic issues facing the world's developed economies over the coming decades, the situation is likely to get far worse before it gets better, even if all forms of debt are stabilized at their current excessively high levels.