An Explanation of the "Chart of the Century": The GDP/Debt Ratio
April 1, 2010
My thoughts on "The Most Important Chart of the Century" - by Christoper Rupe
Recently, this graph has been showing up on dozens of other blogs (here, here, here, here, here, here, here, here for example) and there are continuing questions as to the source of the data as well as the methodology used to calculate it.
The data sources are the Federal Reserve Z.1 Report, specifically, the Total Credit Market Debt figure and the annualized quarterly Nominal GDP number from the Bureau of Economic Analysis. I was sort of the progenitor of resuscitating this graph according to the original Legg Mason format and methodology.
Total Credit Market Debt includes all public debt (federal, state, local) and private debt (mortgages, auto loans, credit cards, etc.). It does NOT include the unfunded public liabilities such as Social Security, Medicare, etc.
Since I have the dataset, and have been compiling and thinking about it for a year and a half, I feel compelled to give my opinion on what I think it means.
The chart is not hard to reproduce, although a bit tedious.
The specific methodology is simply a 4 quarter moving average of (GDP(t) -- GDP(t-4))/(Debt(t)-Debt(t-4)), where 't' is in reference to the quarter in question.
That in and of itself isn't so tedious, however, quarterly data disappears once you go back a few years and all you have left is annual data. So, you have to do a linear interpolation between each of the annual datapoints in order to generate the quarterly data. Incidentally, if you want to get rid of the simple moving average, you of course get a noisier chart with a deeper plunge:
The plunge goes to -0.91.
So, that's really all there is to the method, although there are a number of other ways you can calculate and plot a chart to get the gist of what is occurring here.
For example, Karl Denninger has also done a version of this chart on his own with his "Ponzi Finance Indicator Chart":
Karl's chart has some advantages over the other presentation. For example, I believe Karl's chart uses a year over year % change format for his data rather than a year minus year format. Thus, his chart is not as susceptible to the problem of dividing by very small (close to zero) numbers that the other chart is. Indeed, if you go back in time far enough (I have this data back to 1916) this problem occurs more than once.
If you plot the Legg Mason method over a similar (postwar) time scale it has a look somewhat similar to Karl's chart:
Now, there has been some debate as to the significance of this chart. Some think it is not very significant since it is recognized that if either the numerator OR the denominator goes negative, then the chart goes negative. If both are negative, then the chart yields a positive number. Others think that correlation is not causation w/r/t the impact of debt on GDP. Obviously, some think it is very significant. Nathan Martin referred to it as the 'Chart of the Century'.
I agree with the detractors insofar as this is not a perfect metric of our predicament. GDP, for example, is kind of a fuzzy number. Not just because of tabulation games the government plays along with the constant revisions, but because GDP doesn't even have a universal definition. Different countries calculate GDP in different ways.
I flatly disagree that debt and GDP are not correlated. Indeed, I view this chart as a measure of the Marginal Velocity of Debt in the economy. I have referred to it as such before. In this view, Total Credit Market Debt is synonymous with the Total Money Supply. All money is debt, and this debt has value and has varying degrees of liquidity. To the extent that debt loses it's value or becomes totally illiquid (Home Equity Loans anyone?) it ceases to be money. To the extent that debt is paid back or is defaulted on, money supply shrinks!
This is an unconventional view of money. There would be howls of protest against this view ranging from mainstream (keynesian/monetarist) economists to the austrian economists. Gary North might disclaim me.
However, I have never liked any of the monetary statistics as compiled and named, M0, M1, M2, MZM, etc. I find them too narrowly focused on the liability side of U.S. bank balance sheets. A lot of money is missing in that view. Much of it is obscured by multiplier effects due to the advent of off-balance sheet entities. Some is not counted. Cash held by foreign central banks for example.
The way I see it, all loans must be originated by U.S. banks whether or not they are held on or off balance sheet. This is a distinctly asset side of bank balance sheet view of money. And ALL of this money is accounted for in the Federal Reserve Z.1 as Total Credit Market Debt. So, I have sometimes referred to this statistic as 'Mtotal'.
This velocity measure thus follows the familiar equation of exchange: Velocity = GDP/(Money Supply) and Marginal Velocity = Delta(GDP)/Delta(Money Supply). Not just correlated, but plain old related, as can be seen.
My own view overall is that the chart indicates (however imperfectly) that something significant has occurred. Indeed, nothing like it in 65 years. Anyone can check the Z.1 history and see that Total Credit Market Debt has never had a year over year decline in the postwar era.
Until now.
I view it as confirmatory (in our unique circumstances) that we have 'hit The Wall' with the amount of debt that the economy can carry (debt saturation). Politicians and policy makers have been able to keep the game going by continually and incrementally debauching lending standards.
Think about it this way. A large reason why we have emerged from every postwar recession and renewed the credit cycle to grow to ever greater heights is because the policy response has been to allow, nay, promote the decay of lending standards.
- In the '60s you had to have a 20% downpayment. PMI applied to tiny fraction of mortgage loans. Credit cards required either a security deposit or were secured by a savings account. The point is, they had some kind of collateral.
- In the '70s, PMI grew, especially encouraged by policies like the CRA
- After the early '80s recession, Usury laws began to go away so credit cards eliminated their collateral requirements now that they could charge very high interest rates. Also, the law was changed to take Social Security surpluses and use them to buy government debt. This effectively lowered the lending standards to the government since they now had a guaranteed buyer. Incidentally, this change plus the continued growth in credit set up a feedback loop to artificially lower rates on Treasury securities.
- After the '91 recession, reserve requirements on most bank accounts were eliminated and Adjustable Rate Mortgages began to really grow as a % of home loans. FICO scores were developed in lieu of having any actual collateral for loans. Credit cards were now given away freely to any college kid with no job. Banks begin to financially innovate ways to originate loans but package them into off-balance sheet trusts which are sold, not counting against their capital requirements.
- After the 2001 recession, the gloves come off and anyone can get a loan for anything, no downpayments, bad FICO score, no job. Sometimes the bank will even loan more than the asset is priced at.
We have crashed into the wall.
Marginal Velocity has collapsed.
Christopher Rupe has recently begun to post his thoughts on finance and economics at Financial Minority Report