Monday, April 5, 2010
In case you haven’t noticed, I have become slightly less “optimistic” about the prospects of a sustainable U.S. recovery. I used to think that the household deleveraging story was more of a decade-long project; the economy would cycle throughout. But recent deficit hysteria has me worried; income growth might lapse.
What differentiates this recovery from every other cycle since 1929 is the lingering debt deflationary pressures. There is a very large overhang of U.S. household financial leverage that’s going down one of two ways: the easy way, through nominal income growth, or the hard way, by default. Unfortunately, the hard way is rearing its ugly head.
The chart above illustrates private-sector financial leverage (debt burden). Not a surprise; but, the real leverage problem is in the household sector, and to a much lesser degree, the non-financial business sector. Household debt burden is the ratio of the debt stock (generally mortgages outstanding plus consumer credit) to income flow (personal disposable income), while “de-leveraging” is reducing this debt burden.
Given that the burden has a numerator (debt stock) and a denominator (income), de-leveraging can occur through either variable. As such, I see three (general) de-leveraging scenarios:
1. If there is no income growth, then households must manually pay down debt at the cost of current consumption. The consumption decline drags the economy, and some default results.
2. If income growth is positive, then the degree to which households must pay down debt at the cost of current consumption will depend on the pace of income generation. This is the most macroeconomically-benign scenario.
3. If income growth is negative, i.e., deflation, then real debt burden rises. 30-yr mortgage payments, for example, are fixed in nominal terms and become more difficult to meet as income declines. In this case, widespread default is likely.
Of course, these are just three broad categories, but I believe that my point has been made. McKinsey wrote about this last year. Clearly, choice 2. is optimal. However, evidence is pointing to a de-leveraging process that is more of the 1. and/or 3. type, especially as the federal stimulus effects run dry (although I have noted before that there is room for error in the measurement of the income data).
The chart illustrates annual growth of disposable personal income minus annual growth of disposable personal income less government transfer receipts (DPI – DPIexT). The variable DPIexT proxies the personal income growth currently generated by the private sector only. Note: this is a calculated number, based on the BEA’s monthly personal income report (Excel data here). The spread has never been wider, 2.1% Y/Y DPI growth over DPIexT growth in February, spanning every recession since 1980.
The government is propping up income (as it should be). Spanning February 2009 to February 2010, DPI averaged 1.2% Y/Y growth per month, while DPIexT averaged -1.1% Y/Y per month. Further, since the onset of the recession DPIexT fell an average 0.03% M/M (over the previous month), while DPI grew 0.18% M/M.
The economy has crossed the threshold and is expanding – phew! However, without a burst of export income, it’s going to take a lot more than 123,000 private payroll jobs per month to free the economy of its fiscal crutch. (I debated whether or not to use the term “crutch” when applying it to fiscal policy because fiscal policy is not a crutch; but the metaphor works.)
Households WILL drop leverage further; it’s just a matter of how smoothly.