Some Thoughts on Debt Saturation and Growth

We recently received multiple references to this chart:

This one from Joe Miller (attributed to a David P.) was accompanied by the statement:

Regardless of whether you call it debt saturation or diminishing return on new debt, the notion that taking on more debt will magically enable us to “grow our way out of debt” is not supported by data.

Although we intuitively believe the author is right, we do not think he has proven his point. We discuss the post and explore our thoughts on the topic.

First, let’s discuss Miller’s short post and particularly the graph. The blue line on his graph is the annual change in GDP, the figure the market and the media pay attention to. Although the author states that the purpose of this chart is to examine the relationship of total debt to GDP, the red line shows the ratio of the annual change in total credit market debt owed (TCMDO) to total GDP, as a percentage. It is difficult to extract any useful meaning out of this chart other than debt had been growing at a faster rate than GDP and suddenly it stopped. This does not prove or imply debt saturation.

But let’s assume that TCMDO is a useful measure of total debt in the economy (see the discussion A Note on Credit Market Debt) and see if we can extract something useful. If we want to measure the change in TCMDO then it would seem more appropriate to measure it against the change in GDP:

Figure 1. Annual percentage change in total credit market debt relative to the annual percentage change in GDP.

In this graph, a value of 1 means that debt and GDP are increasing at the same rate. It is only after about 1982 that debt creation outpaced GDP growth and at an accelerating rate up to the start of the last recession when it plunged to values below the rate of GDP growth and in fact contracted (went negative) before turning back up.

The second statement the author makes is: Any gap between the red line and the blue line is what I would call the creation of debt in excess of income. Now he rightly observes that in 2008 there was about a 25% difference. What is perplexing is that he ignores that data that shows at the height of the crisis in 2008-2009, there was little or no difference between the lines. In other words, currently the change in credit market debt is growing at the same rate as GDP. Since gross domestic income is roughly equal to GDP, the author’s assertion is reasonable except that we do not agree that the red line is a proper measure of change in debt.

Debt Saturation

Rather than focus on the entire economy, it might be more fruitful to focus on the household sector which accounts for roughly 70% of GDP. And consumption drives business investment which is another major component of GDP.

Figure 2. Household Credit Market Debt Outstanding relative to Disposable Personal Income.

In Figure 2, we see that personal debt to disposable income rose linearly from the start of data collection in the early 1950s at about 35% to 65% in the mid 1960s. It stayed around that level into the mid 1980s when it began increasing and reached 100% in 2002. Its rate of increase picked up and the ratio peaked at 130% around mid 2007. It has since fallen to 110% with no indication the decrease is finished.

What does debt saturation mean? Unless we can answer this question, we can’t measure it. The intuitive meaning might be the level of debt to income at which more debt cannot be taken on. This obviously various from household to household, but Figure 2 gives us an aggregate for the economy where a saturation point might be measured.

There is also the question of whether there is a point at which the ratio cannot move higher over time. That would be a kind of absolute saturation point. We might also have interim saturation points at which the increase of the ratio halts and possibly retracts, only to move higher later.

We see retractions in the debt/income ratio in the late 1960s of roughly 12% over 5 years, and the early 1980s of roughly 6% over 3 years but in both cases, the increase continued after the pause. The current maximum in the ratio occurred as noted, in mid 2007. The ratio has declines roughly 15% in 5 years since then.

Two possibilities come to mind. At some point in the near future, the ratio may again turn around and move to new highs. If this happens, then the 2007 peak might be considered to be an interim saturation point. On the other hand, if this peak is not exceeded in the future then 130% would be an absolute saturation point.

We realize that the theoretical considerations are more complex. For example, can the economy change in nature over time such that what is a saturation point today may not be approachable in the future or may be easily exceeded in the future? Empirically, we would argue that debt saturation for households occurred in 2007 at 130% of disposable income. Since we are still close to that point, the author’s assertion that we are not in a position to grow our way out of debt (through borrowing-facilitated consumption) would seem reasonable.

Addendum: The Real Effects of Debt

Since we wrote the post we were reminded of a paper by Stephen G Cecchetti, The real effects of debt in which he studies the issue When does debt go from good to bad? Also, we reference an essay by lacy Hunt which we reviewed: The Policy of Doom. His conclusion:

Our results support the view that, beyond a certain level, debt is bad for growth. For government debt, the number is about 85% of GDP. For corporate debt, the threshold is closer to 90%. And for household debt, we report a threshold of around 85% of GDP, although the impact is very imprecisely estimated.

Not only is debt a problem for future growth but the problem is compounded by unfavourable demographics. Lost productivity due to an ageing populations and rising dependency ratios from health and social services have the potential to slow down growth even further.

As the author notes: Without debt, economies cannot grow and macroeconomic volatility would also be greater than desirable. But when debt levels become too high, the benefits reverse and work against the debtor.

Hence, instead of high, stable growth with low, stable inflation, debt can mean disruptive financial cycles in which economies alternate between credit-fuelled booms and default driven busts. And, when the busts are deep enough, the financial system collapses, taking the real economy with it.

The author provides two graphs showing the levels of debt held by the various sectors.

Figure 3. Non-financial sector debt as a percentage of GDP.

Figure 4. Non-financial sector debt, real levels, deflated by consumer prices.

While we find a threshold of 84%, and that the impact of household debt on growth is
first positive and then negative, our estimates lack statistical precision. In fact, the p-values
for the test of whether the coefficients are zero is nearly one half. So, while we may believe
that there is a point beyond which household debt is bad for growth, we are unable to reliably
estimate that point using the historical record available to us.

Finally, we note the author’s comment on the accuracy of the threshold estimate:

While we find a threshold of 84%, and that the impact of household debt on growth is first positive and then negative, our estimates lack statistical precision. In fact, the p-values for the test of whether the coefficients are zero is nearly one half. So, while we may believe that there is a point beyond which household debt is bad for growth, we are unable to reliably estimate that point using the historical record available to us.

In our study of the problem we chose to use disposable income rather than GDP as the basis for estimating a debt tolerance level beyond which the consumer would be negatively impacted. Disposable income is a direct measure whereas GDP is a  much more indirect basis.  Cecchetti’s threshold of 85% debt/GDP would be a floor for the level of impairment. Our ratio of 130% of debt to disposable income suggests an empirical ceiling with the current level of 110% registering severe impairment.

For comparison purposes, we can view household credit market debt to real GDP.

Figure 5. Household Credit Market Debt Outstanding as a percent of real GDP.

At roughly 95%, household debt/GDP is still appreciatively above the 85% threshold of Cecchetti. This supports our thesis that the American consumer is severely debt impaired and will not be a source of significant economic growth for some time to come.

Related Posts

 

Powered by WordPress | Designed by: photography charlottesville va | Thanks to ppc software, penny auction and larry goins