The United States of Debt – Part I

The economy of the United States is highly leveraged. Through the boom times of the 1980s through the 2000s, total US private and public debt to GDP has risen sharply. Since the 2008-2009 recession, private debt has started to fall while public debt has risen, while the country's total indebtedness has remained relatively stable. Most US citizens have the nagging feeling that our debt burden has gotten too high. If we know that debt to GDP needs to come down, however, we don't know to what level it should come. I have decided to analyze this issue in a series of posts called "The United States of Debt".

In this the first installment, I will run through the big picture. The first section discusses how the recent run-up is not surprising in the context of the long cycle, and that we have seen these financial booms before, during the Gilded Age following the Civil War and during the Roaring Twenties. The second section dissects the recent debt buildup between the private, public and financial sectors. The last section discusses why the recent financial boom has been so long lasting and powerful relative to the Gilded Age and Roaring Twenties.

The next three installments will analyze the financial health of the Household, Business and Government sectors and to make policy recommendations on how to promote the financial health of those sectors.

The long wave debt cycle

It's no secret that the United States has a debt problem. This long term chart of total, economy-wide debt to GDP demonstrates it pretty well.

Chart 1

The long term pattern of debt peaks corresponds with the long wave, or Kondratiev Cycle (for a more detailed description of this concept, see here). During the "autumn" phase of the "financial bull", inflation is falling, thus interest rates are falling and asset values are rising. While asset values are rising, the cost of borrowing is falling, which creates a perfect environment for using leverage for investment, whether in the financial markets, real estate markets or for corporate investment.

Eventually, the asset bubble pops, marking the transition to the "winter" phase, or the "real bear". It is at the bottom of the first winter recession that the debt-to-GDP ratio peaks (previously in 1875 and 1933). Why? Because as asset values start to fall, the high level of underlying debt results in a deflationary spiral of forced asset sales and financial failures. Nominal GDP falls while the level of debt remains roughly the same (or rises).

Eventually, the economy bottoms and debt-to-GDP turns down. The decline in debt-to-GDP comes from a combination of rising real GDP, rising inflation (which reduces debt in real terms) and debt destruction trough defaults.

Chart 2

At the beginning of the Great Depression in 1929, the bulk of US debt was corporate debt tied to the investment bubble in auto manufacturing, electric utilities, houshold appliances, radio and other sectors that made up the "new economy" of the Roaring Twenties. Corporate debt to GDP collapsed from over 100% in 1933 to under 30% by the mid-1940s. Household debt fell from 50% of GDP in 1933 to roughly 15% in the same time period. Much of the drop in total debt to GDP came during the 1933-1936 period, when the dollar was devalued by 50% and economic growth was very high (albeit from an extremely low base). The ratio rose again when the economy relapsed in 1937-38, but resumed its decline when the economy recovered modestly in the pre-war years. The end of the long depression years came when government debt to GDP began to expand during World War II in the early 1940s. Government debt to GDP balooned from under 50% to about 100% of GDP during World War II, while private debt to GDP declined.

Total non-financial debt to GDP remained between 125% and 140% from the end of World War II to the early 1980s. During that time, government debt to GDP fell from about 100% to 40%, while private non-financial debt rose from 40% to 100% by 1980. During the same time, financial debt to GDP rose from under 5% to about 20% by 1980, returning to roughly the same level that prevailed in the late 1920s.

The consumer and financial debt explosion since 1980

Table 1

Table 1 above walks us through how our debt load has evolved since the early 1980s. For the sake of cyclical consistency, I show the recession years of 1982, 1990, 2001 and the recent recession trough of Q2 2009. There has been a steady increase in private sector debt, particularly among households and the financial sector. During each business cycle, the private sector tacked on 50-60 percentage points of indebtedness. The government sector, on the other hand, has see-sawed in a one-cycle-up, one-cycle-down pattern.

The largest increase in indebtedness has come from the financial sector, having increased by 93 percentage points between 1982 and 2009. As can been seen in Chart 2, about half of that increase came from the Government Sponsored Entities ("GSEs"), namely Fannie Mae and Freddie Mac. That debt has from a practical standpoint become government debt as well.

Since the 2009 recession has ended, total debt to GDP is has declined by three percentage points. Private sector debt has dropped by 14 percentage points while government debt has risen by 11 percentage points. So far the government has been filling the classic Keynesian deflation-fighting role by borrowing and spending to offset a decline in private borrowing and spending.

The problem with this analysis so far is that while interesting, it doesn't mean a lot without knowing what the optimal debt to GDP level is. Just because debt to GDP used to be 150%, doesn't mean that was the right level. In fact, for all we know, 150% was way too conservative and presented a great opportunity to lever up to the optimal level north of 300%. My gut tells me that the truth lies somewhere in between these two extremes.

The Second Gilded Age

So what happened in the period from 1980 to 2009 that led to an explosion of US indebtedness? We have had a confluence of factors that created the ideal environment for the greatest bull market in the history of the world:

  1. A roughly 90% devaluation of the dollar in nominal terms against gold;
  2. A general policy mix of tight monetary policy and loose fiscal policy;
  3. A decline in long-term interest rates from nearly 15% to 4%;
  4. Tax policies that reward borrowing and equity capital gains; and
  5. Financial deregulation and the expansion of the GSEs.

With the devaluation of the dollar (also known as "inflation"), the denominator in which the S&P 500 is quoted has declined, providing a lift to the nominal value of the S&P. That said, however, over the past 50 years, the price of gold and the S&P have merely kept pace with each other (see Chart 3). Since 1955, 100% of the excess return of the S&P 500 over gold is attributable to dividends.

Chart 3

Source: economy.com, Standard and Poors, tylernewton.com

 

Or looking at it another way, Chart 4 shows the S&P 500 expressed in terms of ounces of gold.

Chart 4

Source: economy.com, Standard and Poors, tylernewton.com

If you ignore the outlying years of 1997 to 2001, the pattern is pretty clear: a steady buildup to about 3 ounces of gold before the plunge to below 1. The only question now is how low the ratio goes. The previous bottom was 0.22 in 1980. That would imply the S&P falling to 265 at today's price of gold, or conversely, gold rising to 4,881 at today's level of S&P.

Nevertheless, since 1955, the price of gold has risen from $35 to $1,205 and the S&P 500 has risen from $35 to $1074. The ratio of S&P 500 to gold seems to spend 15-18 years above 1 and 15-18 years below.

The other factor that drove the bull market in stocks and bonds has been the relentless decline of interest rates from 1980 to 2000. As can be seen in Chart 5 below, since the early 1960s, the earnings yield (the inverse of the price-to-earnings valuation multiple) of the S&P 500 has pretty closely followed the yield of the 10-year treasury bond. Prior to the 1960s, when inflation was very low or even negative, earnings yields were far higher than treasury yields.

Chart 5

Source: economy.com, Standard and Poors, tylernewton.com

As can be seen in Chart 3, in the 1970s, when the value of the dollar plunged and the price of gold rose, stocks did not follow because of the rise in interest rates which provided stiff competition to stocks. As treasury yields rise, the required return from stocks' earnings yield goes up as well. That pattern reversed in 1980. Tighter monetary policy gradually brought down interest rates, but loose fiscal policy kept inflation positive. The combination of falling interest rates and moderately positive inflation is the ideal combination for stock returns.

Now that we are likely shifting to a deflationary environment of falling debt to GDP, we may be seeing a shift to rising earning yields while interest rates remain low. This was the combination that prevailed in the 1930s and 1940s. We are already at a spread between earnings yields and 10-year treasuries not seen since the mid-1950s. If inflation stays as low as it is currently (around 1%), goes lower or even consistently negative, we will very likely see earnings yields back to the level of the early 1950s by the middle to end of this decade.

Rising Asset Values lead to Higher Indebtedness

As asset valuations rose from 1980 to 2008, investors had the opportunity to use financial leverage to supercharge returns. Just witness all the people who made a fortune in real estate over the past thirty years: inflation made the value of rents and real estate rise while financing costs fell. Leveraged buyout practitioners could borrow money and benefit from the combination of rising cash flows and valuation multiples while benefitting from falling interest rates. It's pretty simple. In a thirty-year bull market in asset values for stocks, bonds and real estate, the use of leverage to "get-rich-quicker" is a natural outcome.

The Government has encouraged leveraged speculation

Many of the "supply side" tax policies of the past thirty years have encouraged the use leverage as well. Ostensibly, policies like cuts in upper-end income tax rates, dividend tax rates and capital gains tax rates are meant to encourage savings and investment. In reality, they don't directly encourage savings or investment, they encourage the reward from investment profits. Businesses are the ones that actually invest. If you wanted to directly encourage investment, you would allow for the full deduction of capital expenditures in the year incurred and make the R&D tax credit permanent. If you wanted to directly encourage savings, you would impose consumption taxes or large deductions for money put into investment accounts.

We do, however, directly encourage the use of debt by making interest tax-deductible for businesses and real estate investors, while dividends, on the other hand, are double-taxed at the business and individual level. The government even further subsidized residential real estate interest rates through their "implicit" (now explicit) guarantee on securities issued by the government sponsored entities ("GSEs") Fannie Mae and Freddie Mac.

Private equity and real estate investors had this figured out. Take the subsidy for borrowing money and use it to generate subsidized capital gains profits. Benefit from rising asset values as interest rates fell. And benefit from rising cash flows driven by government-generated inflation. In such an environment, it's frankly a wonder we have as little debt as we do.

I bet this makes you wish you had this all figured out in 1982.

One thought on “The United States of Debt – Part I

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 )

Google+ photo

You are commenting using your Google+ 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.