They May Cancel My Subscription, But…
Epstein, as an economist, makes an adequate journalist. I generally skim his columns for amusement purposes; with plenty of Wall Street economists out there who have even less excuse to be plain wrong as much as they are, it doesn’t usually seem worthwhile to pile on the guy. But over the last few months he has gone on and on and on about something he believes he has figured out that everyone else is oblivious to. He argues that the continuing decline of consumer credit isn’t worrisome at all, because it confuses a stock with a flow. But, says Epstein, we shouldn’t worry about the decline in credit, which can be from paying off loans or defaulting; only the new loans matter because those are the ones being used to buy stuff.
It doesn’t seem to occur to Epstein that while thousands of economists over a period of decades can certainly be wrong, you ought to have a really compelling reason that people don’t get it. In this case, he’s just spewing nonsense.
Consider a person who uses his credit card to buy a new computer for $1,000. The next month, he can do one of two things: he can pay off the credit card, or he can spend that $1,000 (that he would otherwise put toward that balance) on something else. Choosing the latter is clearly the same effect as paying off the credit card and then borrowing $1,000 to buy something new, whereas in the former case, the $1,000 does not get spent but saved. That is the only choice for each dollar: save, or spend? It is easy to see that if there is more saving (or less dis-saving), spending is higher, all else equal. [Note that I am not making the value judgment about whether this is a good habit. My own attitude is more along the lines of this all-time classic skit from Saturday Night Live: Don’t Buy Stuff You Cannot Afford.]
Epstein asserts that we should ignore the retirements of debt and only focus on the new loans. But as I just pointed out, keeping an outstanding balance produces the exact same result that paying the balance off and taking out a new loan does. There is no difference. Epstein gets confused because he thinks of the “debt pay-downers” and the “new borrowers” as different people, which of course they usually are. But the changes in the stock of savings, net of debt, is what matters to consumption…whether it is one person doing the saving and borrowing, or millions.
Now, a decline in credit isn’t bad per se, especially in the long run. Increasing societal savings tends to lower the real cost of capital for productive enterprises (which capital comes of course from savings, since savings is necessarily equal to investment). And a lower real cost of capital for productive enterprises tends to lead to greater growth in the long run…which is, incidentally, why deflation is such a bad thing. So I am not arguing that we shouldn’t be paying down debt. Indeed, if the government sector keeps borrowing then our private borrowing will continue to be crowded out anyway and I think that is partly what is happening. But Epstein’s tortured explanation about how a decline in credit doesn’t imply lower private consumption (all else equal) is just wrong.
Bonds were surprisingly buoyant today. Some of that was due to early rumors that central banks were buying fixed income instruments, but technical factors came into play as well – since bonds have failed to break down, for now at least, some of the bid was doubtless short-covering as TYM futures moved to the highest level since March 24th (+10.5/32nds, 3.81% 10y yields). There may also be some incremental nervousness about the CPI data due out tomorrow; the memory of the negative core print two months ago lingers.
That being said, the fear of another weak core CPI print (the consensus calls for +0.1% on headline, +0.1% on core) doesn’t seem to linger much among TIPS investors, who have pushed 10y breakevens back above 2.30% and 10y inflation near 2.70%. One-year CPI swaps are at 1.30%, near the highs of the last year (see Chart below) and probably giving some credit to the possibility of a lower near-term core print.
Personally, I think the consensus estimates for CPI are about right, with risks symmetrical around the core at +0.1%. Those prints would make the year-on-year CPI rise to +2.4% from 2.1% and the core to drop to 1.2%, a new cyclical low and only a bit above the 44-year lows at 1.1% (reached in 2003 during the deflation scare). Of course, conditions this time are very different, with most of the deflation being caused by the unwinding of an asset bubble in housing, but lower readings in core CPI will give the doves on the FRB some further excuse to keep policy loose for longer. The disinflationary tendencies to core, even including housing, are exaggerated at the moment but I still don’t expect the bottom in core CPI to come until early Q4.
Retail Sales (Consensus: +1.2%/+0.5% ex-auto) is also due out tomorrow; the Beige Book and more Bernanke testimony will provide a little more grist for the mill of Wednesday’s trading. I remain bearish on bonds.