Deleveraging, Deceleration and the Double Dip
Much optimism flowed from last week’s declaration by the National Bureau of Economic Research that the US recession officially ended in June 2009. How nice of them to let us know.
Markets reacted warmly and the 8 per cent rally in US stocks through September seemed more important than the revelation that the US Fed is worried enough about deflation to be planning another round of quantitative easing – dubbed ‘QE 2′.
I wish I could share the market’s (and the NBER’s) irrational exuberance, but the key indicator that explains where the US economy and its already disastrous employment situation is headed implies that even QE2 won’t set the US on a course to renewed prosperity.
It’s also important to note that the ‘strong’ Q2 US earnings figures seen in July are partly the result of dramatic cost-cutting in US firms – a nice way of saying mass lay-offs. Nonetheless the stronger bottom lines keep producing market exuberance, even if the factors behind those bottom lines will lead to future deflation rather than a boom. As Forbes writer Joshua Brown puts it “there is an effervescence in the air as we head into the Q3 reporting period (starting October 7)”. He thinks the market will once again turn south, and based on that pesky fundamental called the real economy, it should.
So should it on the basis of the key indicator that explains the origins of the apparent stabilisation that led the NBER to declare that the recession was over in June 2009.
For a long time I’ve focused on the contribution that the change in debt makes to aggregate demand, in the relation that “aggregate demand equals the sum of GDP plus the change in debt”. An obvious extension of that was that “change in aggregate demand equals change in GDP plus acceleration in the level of debt”—which would imply that change in unemployment is driven by changes in the rate of growth of debt.
Though I was aware of this implication of my analysis, I held off from testing it because I was concerned that this was pushing the data one step too far.
A physical system with a similar relationship between velocity (the rate of change of one variable) and acceleration (whether the velocity of another variable is increasing or decreasing) would generate a large volume of sufficiently detailed data that the relationship could be empirically tested.
But the economic system, with the large time lags in data collection, survey methods rather than direct measurement, and the dodgy practices statisticians are forced into by politicians and economic bureaucrats who often don’t want raw information to be available? I just thought that the relationship, even though it made sense, wouldn’t be discernible from published statistics. So I held off.
It turns out that I shouldn’t have been so cautious: the data well and truly supports this, on the surface, weird causal relation: the change in employment is strongly affected by the acceleration or deceleration of debt. This can give the paradoxical result that the level of employment can rise, even when the economy is deleveraging, if the rate of deleveraging slows. This phenomenon has driven the apparent stabilisation of the US unemployment rate (though of course the more meaningful U-6 measure has risen to 17 percent, and Shadowstats puts the actual unemployment level at 22.5 percent–well and truly in Depression territory), and it is highly unlikely that it will last.
My uncharacteristic timidity means that I have to doff my cap in the direction of the three economists who first published on this topic: Biggs, Mayer and Pick. They first showed the correlation between what they called “the credit impulse”—the rate of change of the rate of change of debt, divided by GDP—and both GDP and employment (for those who have access to research from Deutsche Securities, they have a simpler explanation of their analysis in Global Macro Issues for December 17 2009: “The myth of the credit-less recovery”).
The chart below shows my confirmation of the relationship with the data on the annual change in unemployment in the USA and the annual rate of acceleration of private debt since 1955. The correlation is -0.67: a staggering correlation of a first and a second order variable over such a period, and across both booms and busts.
The two dotted red lines labelled “S” and “E” show when the NBER thinks this recession started and ended, and they neatly coincide with turning points in the credit impulse—an indicator that the NBER is not even aware of, let alone one that it considers when attempting to date recessions.
Superficially, one might think that since the credit impulse does indicate when unemployment is going to rise or fall, then the current data implies that the recession is indeed over—even if the NBER doesn’t understand the actual causal dynamics at play.
But the chart also shows that there has never been a turndown in credit like this one—the peak rate of deceleration of debt was over 25 percent, versus a mere minus 6 percent in the deep recession of the 1970s. And though the rate of acceleration of debt has the most direct impact on employment, ultimately all three factors—the level of debt (compared to GDP), its rate of change, and whether that rate of change is increasing or decreasing—must be taken into account.
It’s complicated, so an analogy with driving makes it easier to comprehend.
Consider a drive from Los Angeles to some destination East (if you’re an Australian reader, consider a drive West from Sydney), where the drive out represents increasing debt, and the drive back home represents falling debt.
The level of debt compared to GDP is like the distance to be travelled, and today the US has a lot further to travel than it did in the 1950s: 5 times as far, in fact. It’s like the difference between a drive to New York and back, versus a return trip to Utah.
The rate of change of debt (with respect to GDP) is like your speed of travel—the faster you drive, the sooner you’ll get there—but there’s a twist. On the way up, increasing debt makes the journey more pleasant—the additional spending increases aggregate demand—and this experience is what fooled neoclassical economists (who ignore the role of debt) into believing in “the Great Moderation”. But it increases the distance you have to travel when you want to reduce debt, which is what the USA is now doing. So it’s great when you’re driving from LA East (increasing debt), but lousy when you want to head home again (and reduce debt).
With that far to travel back home, you might be tempted to accelerate—which is akin to increasing the rate of change of the rate of change of debt (it’s a measure of the g-forces, so to speak, when they can be generated by either rapid acceleration or rapid deceleration). Acceleration in the debt level when it was rising again felt great on the way out: booms in the Ponzi Economy the US has become were driven by accelerations in the rate of growth of debt. Equally, acceleration in the opposite direction feels dreadful: as the rate of decline of debt increases, aggregate demand collapses and unemployment explodes.
What actually feels better in the reverse direction is deceleration—reducing the rate at which debt is falling—and that’s what’s been happening in the last year.
But here’s the problem: too much deceleration and you actually reverse direction: you start heading East again, rather than returning home. That wouldn’t be a problem if all you’d done was drive to Utah, but instead you’ve hit New York instead: drive any further, and you’re in the Atlantic.
With the level of debt the USA has accumulated, the prospect that any sector of it (apart from the government) can be enticed to go back into accumulating debt once again is remote. So the deceleration in the rate of reduction of debt that is occurring right now will ultimately give way to at best a constant rate of decline of debt, and at worst another acceleration—and the dreaded “double dip”.
These next two quarterly charts emphasise the dilemma: the stabilization in employment has occurred because the rate of deleveraging has slowed, which registers as a positive in the “rate of change of the rate of change”:
But the rate of change of debt is still negative: it’s just risen from a low of -6% to -2%. For the deceleration effect to continue, the US debt car would need to stop its return drive from New York to LA, and head back towards New York once more: the level of debt relative to GDP would need to rise.
This is highly unlikely when all sectors of the American economy bar one (non-financial business) are already carrying more debt than they were in the depths of the Great Depression, when the debt ratio had been driven higher by deflation.
So the deceleration in deleveraging should give way again at some point, and then the NBER may be forced to begin dating the next recession—which is still a continuation of the current Depression.