Matt Rognlie has two important posts – Deleveraging and monetary policy, balance sheet and reality– arguing against the idea that deleveraging is a special problem for the economy (“Few popular terms irritate me as much as ‘deleveraging'”). You should read both posts. Rognlie is arguing that there is nothing interesting in the purple housing and debt circle, or at least nothing interesting that can’t be solved by monetary policy. There’s a market that isn’t balanced and we just need to bring the price to clear.
Now this is different from arguments that there is debt hangover problem but fiscal and/or monetary policy – getting people jobs and better wages – is the best way to deal with it. This argument would say focusing on debts first is the wrong end to grab the stick. And looking at savings rates versus weak income growth it’s not obvious that savings rates are the drivers in keeping consumption in check. See Christina Romer’s recent editorial here:
…evidence suggests that high debt is holding back consumer demand. But it doesn’t follow that the government needs to directly lower debt burdens to stimulate job growth…
History actually suggests that the “tackle housing first” crowd may have the direction of causation backwards. In the recovery from the Great Depression, economic growth, which raised incomes and asset prices, played a big role in lowering debt burdens. I strongly suspect that fiscal stimulus will be more cost effective at speeding deleveraging and recovery than government-paid policies aimed directly at reducing debt.
I will say some meta-things in support of Rognlie before listing the counter-arguments. I have seen way too many people invoke this argument and Reinhart and Rogoff’s “This Time It’s Different” as a way of absolving responsibility for taking on the crisis. This is both inside the administration (Secretary Geithner, President Obama) and outside. Indeed saying a version of “Obama didn’t take ‘Reinhart/Rogoff’ seriously” is becoming a go-to critique on the Right, even though the causation problems aren’t clear in the RR thesis at all. Indeed Reinhart/Rogoff can be read just as much as an indictment of not acting than trying to act.
Secondly, let’s take heed to not go out of our way to make up problems to solve. Aggregate demand is weak. Okun’s law is holding since the 1980s, the BEA’s update showed us the economy had much more of a hit to GDP than we thought, structural unemployment as viewed through the Beveridge Curve is on target for what we expect, etc. Let’s be clear on what we need explanations to solve.
That said, I have some disagreements. First off, he argues that “In the aggregate, the assets of American households are still far higher than their liabilities” and implies that it is tough to see why the current level of liabilities would or wouldn’t be sustainable.
I haven’t updated this graph (from Fed Flow of Funds) but the relationship continues:
Short-term interest rates are at zero, yet consumers are still deleveraging. Think that through – debt is “free” but consumers want less of it. That defies normal economics and any type of regular good unless you start to bring balance-sheet issues.
But can’t some level of monetary policy fix this? Rognlie would say that debt is not free and interest rates can in fact go lower through operations like QE. The future path of interest rates can be communicated by the Fed which will push rates as well. I think these are a fantastic ideas, if only on the corporate side, but I’m not sure it hurdles the fence entirely.
Again, what is driving what here. As Henry Kaspar argues:
Excess indebtedness forces debtors to reduce expenditures in order to be able to service their debts – i.e., it forces a transfer from debtors to creditors. The creditors’ propensity to spend is lower than that of debtors; after all this is why they are creditors. They would want to lend on the means transferred to them, but do not find enough households or firms willing and able to take on more debt. To use Brad DeLong’s term: there is a lack of safe assets…
Because of the lack of safe assets creditors hoard money… But money hoarding is a symptom of the crisis, not its cause. Put differently, causality goes from lack of aggregate demand to money hoarding, not the other way round.
Rognlie also brings in the idea that for every debtor is a creditor, and if consumers are paying down debts someone else is getting a windfall and that should balance.
But let’s take an example I’ve been thinking about. Let’s say you sell your house for $100,000, but your mortgage and closing costs are $110,000. As such, you bring $10,000 to the closing to give to the bank that held the mortgage. What are the economic consequences of that? Is that the same than if you bought a $10,000 car? Spent an extra $200 every weekend on leisure activities?
If the bank can’t find profitable opportunities for loans, which by all accounts it can’t, then it isn’t the same. And if the bank has a weak balance sheet – let’s pretend it is Bank of America, which needs as much cash as it can get – then it also doesn’t get lent out. So that’s a problem.
And to the extent that monetary policy works through housing then we have some other major problems to deal with. As QE architect Joe Gagnon has argued, QE was blunted by the inability for underwater homeowners to refinance. The threat imposed by an addition 2-3 million foreclosures in 2012 would make any rational homeowner wary. Indeed a simple lesson of combining finance and macroeconomics is to never force the mass selling of assets into a weak market where there are balance-sheet feedback loops, especially if you can keep productive uses going. And unless hosting a crackhouse is productive, the “shadow inventory” of abandoned housing is just mucking up the normal building cycle. Securitization fraud is a mess of conflicting legal liabilities that are being fought through the courts and the AGs right now, which also hurts the market.
Now the link between debt and low employment is just that, a link. The causation isn’t specified, but evidence on foreclosures gets us closer. Here’s Mian and Sufi with the best empirical evidence there is:
We also find much sharper drops in employment, both during and after the recession, in counties with high household leverage. The theoretical links between leverage and employment do not yield an obvious prediction. First, we would expect over-levered households to supply more labor in order to pay off their debts. Second, employment in a given county is not directly linked to consumption in that county, given that the factors of production are often outside of the area. Despite these issues, we find that employment in high household leverage counties dropped by 8 percent from 2008 to 2009 and remained depressed through the end of 2010…
The recent financial crisis has led to almost 3 million U.S. households going into foreclosure, and the number is expected to increase….By comparing states with different legal requirements on foreclosures, we find that state laws have a large impact on the incidence of foreclosures. We find that foreclosures have large price and real effects. From 2007 to 2009, foreclosures were responsible for 20 to 30 percent of the decline in house prices, 15 to 25 percent of the decline in residential investment, and 20 to 35 percent of the decline in auto sales.
So I think there’s something here, but I think it is important to understand it doesn’t negate the other solutions. What’s your take?
Rortybomb 
What I don’t understand about the deleveraging/balance-sheet-recession argument is that there’s so little evidence for it in the data. The personal savings rate is not high by non-bubble standards, which means that Americans have not, in fact, made unusually large cuts to their spending. Indeed, the negative wealth effect from the bursting of the housing bubble and the stock market’s decline can explain almost all of the drop in spending that we’ve seen since the end of 2007. Similarly, Americans’ debt payments as a percentage of income are not especially high — in fact, they’re as low as they’ve been since the early 1990s. That hardly suggests an economy that’s being held back by a massive debt overhang.
The fact that Americans are not increasing their debt load — which your flow-of-funds graph shows – does suggest that weak balance sheets are having an impact on borrowing, but the implication of your argument that Americans should be adding more and more debt is peculiar: given that the savings rate is only 5%, you’re implicitly arguing that a healthy U.S. economy needs Americans to save at abnormally low rates. That just seems wrong in many different ways.
I’d also add that the foreclosure paper doesn’t necessarily demonstrate what you think it demonstrates. We would expect that counties with very high foreclosure rates would also be counties where there was massive overbuilding, and where demand for new homes would have plummeted, devastating the construction industry. On top of this, the negative wealth effect from falling home prices — which, again, is different from the deleveraging argument — would be expected to drive consumption down, with all the concomitant ripple effects on employment and income.
I think that if you look at Americans’ debt payment as a percentage of income, you also have to look at the distribution of those debt payments among income levels. It’s one thing if higher income individuals are paying down high-interest debts, while lower-income individuals are unable deleverage, because low-income individuals are more likely to spend a greater percentage of their income anyway.
It’s also one thing to say that the debt payments as a percentage of income isn’t high, but if the reason why that’s true is because they aren’t making payments on their debts, then that’s a whole other problem. And that’s also a part of the picture.
I’ll blog this next week, but in email conversations with Arpit Gupta I think we found a good way to distinguish the space within the overlap between the Fiscal and Overhang circles. Richard Koo thinks all monetary policy is useless. If the Fed’s rate was 3% reducing it to 2% would make virtually no difference.
Whereas the Krugman paper and others thinks deleveraging is more likely to bring interest rates to clear below zero, at which point the Fed is (supposedly) constrained. (Rognlie – who does not believe the Fed is constrained – brings up in his posts how the liquidity trap is necessarily in Krugman’s paper title).
That explains the bizarre Koo attack on Krugman. It also explains why people like Krugman do not want the Fed to raise rates and why conservatives Koo-style economists who have a model of labor productivity and deleveraging (say Rajan) want to raise rates and others seem indifferent.
I think Romer is closer to the nub of things. A whole bunch of people are unemployed; a whole bunch of employed people have no security that they will remain employed, and they have seen their incomes stagnate or fall while there has been significant cost-push inflation on gas prices.
The role of mortgage debt, I think, is not that the level of debt is unsustainable or a significantly higher percentage of income then before; but that because home prices have deflated so many people are trapped in their underwater mortgages that the economy can’t evolve and innovate properly. If you are stuck in a dead-end job and see opportunity elsewhere, you still might have no choice but to keep driving into the dead end until you crash fatalistically into the wall you see coming up. You’re trapped. No opportunity; no hope; no reason for optimism or expanded household spending.
Also, the frighteningly massive unemployment means that nobody has any bargaining power. Most people have no way of bargaining to steer a larger share of record corporate profits into wage and salary increases. So they know the expected trend for their own income is downward relative to output and prices. That’s depressing, and surely inhibits consumption.
Unemployment and income loss were effects of the financial events which caused the initial crash. But employment pathology and income dearth are now the preeminent causes of the lingering sluggishness and misery, not just its effects. Give people jobs; boost their incomes; convince them that the increases are permanent and not mere one-time stimulatory whacks. That will get us out of the mess.
Also, even if higher debt is not the primary cause of the problem, it is still a very significant household cost, and so cancelling debt would be an excellent, relatively quick and efficient way to boost disposable income.
The necessary income transfer and jobs program can be accomplished through a combination of pure monetary financing, and taxes on the surplus savings of the very wealthy. Note that even the former requires Congressional action. The central bank can only inject money directly into the financial system, or buy up added government debt that must be approved by Congress given the debt ceiling constraints. The Fed cannot hire workers and send people checks. Only Congress can do that. So there is no avoiding the need for stepped up hardball in the political spheres.
I find it dismaying that so many progressives and liberals these days are looking desperately to the monetarists for solutions. The monetarists seem to think everything from photosynthesis to gamma ray bursts are “everywhere and always a monetary phenomenon:; and they also think the central bank has effective control over monetary phenomena. But they are dead wrong about both points. They spectacularly overestimate the role and importance of the central bank in our economic life. There is no central bank solution to this crisis. Leave Milton Friedman and his monetary autism in the grave, and move on.
I can understand why so many progressives and liberals are succumbing to this monetarist seduction. They see the fiscal capacities of government are irredeemably broken, and have given up on politics. Thus they turn their lonely eyes to Ben Bernanke and the Fed governors for salvation, and adopt the irrational quasi-theological doctrine of monetarism, and its ant-fiscalist, laissez faire corollaries, to rationalize their pathetic hopes.
It’s no use. Without a powerful and economically active government sector, our economy will continue to sicken. If the taxing, transferring, regulating, mandating and spending powers of government remain broken, our society and economy will remain broken. We have no option but to take up the fight against the traitorous, stupid and malevolent do-nothing Congress, and to find a way to mount the political pressure that can force them to act.
I don’t really want to nit-pick, because I’m mostly in agreement with you, and maybe this wasn’t your argument, but the idea that debt is “free” to consumers because short-term interest rates are at zero is misguided. Because of the step in the chain consumers are, interest rates charged to consumers are positive in both nominal and real terms.
Just because banks get near-zero interest rates doesn’t mean that they turn around and give those interest rates to consumers. Banks can get free money, but even a 30-year mortgage will be lent at 4% interest.
The simple fact of the matter is that we’re still dealing with a banking system that went through a financial crisis, and instead of dispersing market power from larger players to smaller ones, ended up consolidating market power even more. Thus, the ability of those larger banks to exercise their market power has not been reigned in, either by regulation or competition.
I think there’s something to be said that if we balanced the market and brought the price to clear in the debt markets, the economy would begin to recover and aggregate demand could take off. But that would be because everyone had already hit bottom, taken all the losses they would take, and had gone bankrupt. Sure, not everyone would go bankrupt, but an awful lot would realize losses, and that’s a lot of pain to force people to take. Banks would collapse, people and corporations would go bankrupt, and housing values would plummet. Aggregate wealth would decrease in ways not seen since the Great Depression. And then we could get back on track.
There are other ways of dealing with the problem. We could attack the debt problem directly, or with fiscal/monetary policy. But to say there is nothing interesting about it is to ignore one of the root causes of the financial crisis in the first place.
Thanks for the helpful comments. Patrick, agreed – I’m mimicking Koo’s Holy Grail argument on the free part.
Because it’s come up several times, we looked at the cross-distribution of the growth of debt. 90-100% percentile saw a decrease actually over the 2000s. Biggest hits in the middle of the distribution:
Pingback: Incrementally reducing your negative equity is not saving « Synthetic Assets
Pingback: Incrementally reducing your negative equity is not saving — Clearing and Settlement
Pingback: The Problem’s Not (Monetary) Policy, it’s Politics « thecurrentmoment
I have not had time to fully go over Matt’s post, but his analysis is based on a two period data set 2006 and 2011, does not look at the distribution of assets relative to liabilities over the population, assumes identities are in harmony, ignores historical relationships in terms of debt to GDP, ignores imbalances between consumption and production and hence savings and investment, ignores the fact that excess money supply growth over and above nominal GDP went into asset price inflation pre 2007 and assumes that assets are even now priced correctly.
If GDP has been inflated by debt (for various reasons), and asset prices and supply by excess money supply, then it is possible that a) asset prices have moved above the ability of historical GDP to finance, b) that future GDP growth will be lower in real terms than historical GDP and c) that asset prices will be overvalued.
While an asset value can decline, a debt cannot, unless devalued by inflation or adjusted by default. So if the amount of debt in an economy exposes future growth to higher volatility, or indeed lower real growth, then it matters not that assets exceed liabilities at a point in time: debt is too high and needs to fall, and if inflation cannot adjust, then asset prices and debt must.
Obviously, this is more complex than this, but there is only no problem if the identities hold, that consumption, production, savings and investment are in harmony and debt can be financed and that asset prices reflect real demand and supply relationships in the economy.
Pingback: Consumer Deleveraging Is A Good Thing | The Pacific Bull Moose