Saturday, July 3, 2010

Eroding a Mountain Range of Debt

The above data shows the debt outstanding of various sectors in the US economy.  It comes from the Federal Reserve Z1 release, and the data are annual except for the 2010 point, which is for Q1 (on the graph, the annual data are shown at the mid-point of the year, and the Q1 point is shown 1/8 of the way into the year, so slopes should still be accurate).

My operating assumption is that the main current problem with the US economy is Too Much Debt in the private sector, and that all will not be well until both the household and financial sectors have deleveraged back down at least to something like the levels of the 1990s (at a rough guess).  On the pace so far, it appears likely that that will take at least a decade.  I don't think any of us should be planning on living in a great economy any time soon.

At the moment, there is a great debate occurring between Keynesians such as Paul Krugman, who argue the government should run deficits to stimulate the economy, and the forces of fiscal austerity who argue the government needs to reduce its debt.

In thinking about this, it seems like the key fact is the accounting identity that

Private Sector Balance + Government Balance + Foreign Balance = 0

Now, there are deep structural reasons why the US is running a current account deficit that cannot be changed soon (dependence on foreign oil, offshoring of labor-intensive work).  Developments in Europe are making it even more difficult for the US to run a current account surplus.  So the third term in the accounting identity pretty much has to be positive (foreigners are net lenders to the US).  If we want the private sector to deleverage, then the private sector balance should be positive.  That means the government balance must be negative (the government should be a net borrower).  This is the argument against government fiscal austerity, and it's quite compelling.

At the same time, I think there's a serious magnitude problem with the idea that the US government should run big deficits until the private sector is sufficiently deleveraged.  The private sector is much larger than the government sector, and its debts are too big.  Here's a stacked graph of the private sector debts outstanding:

If the government runs sufficiently large deficits for long enough for this to get back down to, let's say, the level of 1995, the government is going to end up with a debt to GDP ratio approaching that of Japan, and I would say it's politically inconceivable that the US government could sustain that level of debt.

My conclusion is that the private sector deleveraging process is not, and should not, be mainly a "pay down the debt" process.  That appears to me to be impossible.  It needs to be mainly a "write off the debt" process.  Households, businesses, and financial firms with weak balance sheets will need to go through the appropriate restructuring processes - whether it is short sales on houses, principal renegotiations on mortgates,  debt restructurings and bankruptcy of individuals and firms.  Obviously, there will be lots of nasty feedbacks as we go through this process: household debts being written down make the balance sheets of financial firms worse, financial firms shrinking and laying off workers reduce demand for the output of other businesses and worsen their cashflow, etc.

The main role of the government should thus not be "borrower of last resort", but rather "bankruptcy trustee of last resort".  Clearly, in this kind of environment, the largest financial firms could become at risk again.  In that scenario, it's critical that the government is in a credible position to take them over and conduct a restructuring, and that requires that the government's own balance sheet is in decent shape.  Therefore, I don't think the government should either try to massively increase its own debt, or massively reduce it.  I think it should aim for approximate balance, and in the meantime should place its focus on improving the speed, efficiency, and fairness of the various processes for resolving excessive private sector leverage.  I think given the scale of what needs to happen, we should try to find ways to make the processes more compassionate for the individuals involved.

Fundamentally, we've spent the last two decades borrowing and lending far too much, and there now isn't any painless path back to a healthy situation.  We need to get rid of a lot of this debt, and we need to get a lot less comfortable with borrowing as a solution to our problems.

As a side effect of this, it seems to me that many classes of assets are still very much over-priced.  Houses and stocks are likely to be in an overall bear market for as long as the deleveraging process is going on.

It's not going to be any fun.


Black Lizard said...

Households, businesses, and financial firms with weak balance sheets will need to go through the appropriate restructuring processes - whether it is short sales on houses, principal renegotiations on mortgates, debt restructurings and bankruptcy of individuals and firms. Obviously, there will be lots of nasty feedbacks....

You've become a liquidationist!

Black Lizard said...

If one believes in a singularity-related productivity burst, there is another option. Increased productivity is deflationary. Even without that we are seeing deflationary forces build over the last few months (not yoy yet). This means that there could well be room for the Fed to buy more of the debt Treasury issues. If such forces are a long term trend, that debt may well mature while on the Fed's balance sheet without touching public hands.

This would mean essentially that a decent chunk of debt load is financecd by part of the productivity increase... the government gets it instead of gains accruing to savers as the value of money increases. If a huge productivity-related wave of deflation is headed our way, textbook monetarism shows how those gains could be used for society as a whole.

rks said...

and society as a whole has an urgent need to utilize those gains to do stuff to decouple from oil: electrify transport, heating, etc, and build extra electricity generation that will be cheap to run (e.g. nuclear).

Burk Braun said...

Hi, Stuart-

What is "politically inconceivable" is entirely dependent on our level of knowledge, insight, leadership, and imagination. One would have thought decades ago that it would be "politically inconceivable" for us to be importing planet-killing amounts of oil from countries that hate our guts and had brought our economy to its knees once before. But here we are.

Japan has no problem sustaining that level of debt, and we can as well. It needs to be seen as a massive savings program, completely safe and supported by minimal payments of interest by the government (while also being relatively unproductive and paying much lower returns than productive investments can). And, of course, there is no real reason for the government to even incur all this debt in the first place- that is just a service to the rich and to the money markets as a liquidity and savings vehicle. The government could straight-forwardly print and spend money until the tendencies of inflation and deflation were once again reasonably balanced, which would happen when employment had regained a normal balance.

I don't really know where you get your conclusion from ... the accounting equality you put up is quite correct. Thus if the trade balance is persistently negative, and the private balance has gotten way overboard on debt in the last decade+ and needs to retrench, then the only way this is going to work is by government taking up the slack with negative balances- i.e. high net spending. Private debt is far, far more unstable than public debt in the state's currency of issue. All the state has to keep an eye on is inflation.

It seems important to learn the proper lessons from the depression era. One of them is, quite simply, that profligate spending for world war 2 was hugely beneficial- the US and Germany being two great examples. Another lesson is that government employment programs work- they can accomplish great communal projects while employing lots of people and stabilizing the economy generally. A third is that inflation is deadly- it certainly needs to be kept under close watch. But deflation is even more deadly, so nursing deflation as we are now to keep the way over-hyped specter of inflation at bay is a huge mistake.

A fourth might be that the less time spent deleveraging, the better for all concerned, so the best policy is to address deflation aggressively with stimulus spending, move some of the net debt to the government balance sheet, bring down unemployment as soon as possible, and regain an even economic keel as soon as possible. The level of private debt we have to deleverage to is entirely dependent on the general confidence level of the economy. Going on some huge bender in the opposite direction now is not good for us either- it would be like yo-yo dieting.

Anyhow, congrats on telecommuting.. I love it.

Rob said...

For people underwater on their homes (owe more than the market price), why don't we have the banks declare a one-time "mortgage amnesty" where the loan is cut in half say, or lowered to current market price. The half-way route would be simpler to administer since you wouldn't need a third party to estimate the market value of each house. Instead of bailing out the banks, I don't understand why we don't bailout the underwater homeowners directly. Kicking people out of their homes, often homes that they have already paid a lot of equity into, because they lost their job and now cannot keep up with the mortgage payments is somewhat unfair (e.g., none of us can predict that we'll have our jobs for the next future ten years) and anyways will cause a lot of unnecessary hurt for a "lost generation" of families and be a bit corrosive and disruptive to society in general, which is not overall in our country's best interest.

MisterMoose said...

All debt is not equal. If the federal government or a private company or an individual wanted to borrow money to invest in a new computer which would improve productivity and thus help make (or save) more money in the long run, that would be a good investment. If the federal government or a private company or an individual wanted to borrow money to fund a lavish lifestyle that would otherwise be unsustainable, that would be another story...

If Krugman wants to borrow another trillion to fund unemployment extensions or keep teachers on the payroll, that would not help solve the problem. If we borrowed a trillion to, say, electrify the transcontinental railroad system, that might be worth doing.

Michael Cain said...

Your last graph sums it up nicely: the financial sector has been allowed to take on an insane level of debt, and it will never generate the profits needed to pay it off. There is a staggering amount of worthless paper out there, and no one really knows who is holding it.

The federal government would have been much better served if it had taken the $700B or so it used to try to save the financial giants, and set up seven regional banks each with $100B in capital, charged with making sane loans in the "real" economy. And if the Federal Reserve had overseen a relatively graceful liquidation of the giant financial firms rather than buying a trillion dollars of quite probably worthless paper from them.

Stuart Staniford said...


I think the general issue is that a lot of banks will be clearly visibly insolvent if they write down that many mortgages that far.