econcrit

Saturday, February 10, 2018

And the "institutional changes" that enabled the rise of household debt did no more than keep the growth of debt on trend.

I don't want you to get the wrong impression. I think debt is the problem. More than that, I think private debt is the problem, and public debt is a consequence. But it just kills me when people I should agree with speak carelessly about debt.

Couple days ago I interpreted somebody interpreting Krugman's view, and I said: The core of the problem, as Krugman sees it, is the rise of household debt. And I looked at Bezemer and Hudson saying that credit "was no longer good for growth" by the 1990s, mostly because

credit was extended increasingly to households, not business.

When I read stuff like that, I figure household debt must have been growing faster and faster, and that's how we got in trouble. But that is not what happened. I looked at a few graphs, and that is not it.

The black line on these graphs is an exponential trend line. It shows a constant growth rate. It shows "same and same", not "faster and faster". Above the black line is "faster". Rising above that line and curving away from it would be "faster and faster".

The blue line is Household debt, from the fourth quarter 1951 to the third quarter 2017. The red line shows the data used to determine the exponential trend on each graph. The graphs show the trend thru 1960Q1, thru 1970Q1, thru 1980Q1, etc.

Household debt since around 1970 has been well below the trend established in the 1950s:

Graph #1: The Red Line Ends in 1960 Q1. Point 78 on the X Axis is 1971 Q1

Household debt pretty much remained on the trend of the 1950s and '60s, until the mid 1980s. After the early 1990s, household debt ran a little below trend:

Graph #2: The Red Line Ends in 1970 Q1. Point 144 on the X Axis is 1987 Q3

Household debt closely followed its 1951-1980 Q1 trend, right up to the crisis and 2009 recession. This is the only one of these graphs that shows household debt going above trend after 2000:

Graph #3: The Red Line Ends in 1980 Q1.

Household debt going above trend in the 1980s (see Graph #3) moved the trend line up a little. Based on the trend thru Q1 1990, household debt since the mid-1990s has been below trend:

Graph #4: The Red Line Ends in 1990 Q1

Debt running below trend in the 1990s moved the trend line down a little. Be that as it may, household debt after Q1 2000 remains below the 1951-2000Q1 trend:

Graph #5: The Red Line Ends in 2000 Q1

Household debt did a remarkably good job of staying on an exponential growth path right up until the 2009 recession. Note that debt was below trend since the mid-1990s.

If you want to point out that exponential debt can't continue forever, well yeah: We watched it fail to continue, ten years ago. But the problem is not that household debt was growing faster and faster: Debt was essentially on or below trend since the 1950s.

Yes, it was still too much household debt. Yes, there is too much debt everywhere, not just with households. But each of these trend lines shows a constant rate of growth. So if we have to say there is too much household debt now, we also have to say debt growth was a problem right from the start: It was unsustainable, right from the start.

If people make careless claims about the growth of debt in recent decades, they are probably careless with other facts as well. So now the main idea, that rising household debt is the problem, this whole idea has to come into question.

All debt is a problem, even when it is debt we need.

Friday, February 9, 2018

Debt grows faster than GDP

Debt grows faster than GDP. So if you look at the Federal debt relative to GDP, it goes up. If you look at Household debt relative to GDP, it goes up. If you look at Household Mortgage debt relative to GDP, it goes up. Pretty much any debt you look at, if you look at it relative to GDP, it goes up.

So what you can do is, no matter who you want to blame, you make a graph of their debt relative to GDP. And you show people that it goes up, and they say "Yeah, look at that." And then that debt becomes a problem. Or, at least they think it's a problem.

It may be a problem, or it may not. Me, I always say debt is a problem. But here's the thing. If you only look at one type of debt, like Federal debt or Household debt, and you always look at it "relative to GDP", then you will always see it going up. So it doesn't tell you anything.

I like to look at debt relative to "All Sectors" debt. All debt. This way I can see if the Federal debt is growing as a share of all debt, or if household debt is growing as a share of all debt. I can check any component of the "all debt" total. Looking at it this way, if some of them are growing as a share of total debt, then other ones must be shrinking as a share of total debt.

The shrinkers could still be a problem, of course. The shrinking doesn't mean it's not a problem. But the ones that are shrinking are growing more slowly than the ones that are rising. And that could be a useful thing to know.

For example, here is Household debt:

Graph #1: Household Debt as % of GDP (blue) and as % of All Sectors Debt (red)

You have probably seen the blue one before, Household Debt to GDP. Flat in the 1970s, rising since the '80s, and peaking just before 2010. Enough debt to make you shudder.

The red one is the same debt, relative to All Sectors debt. This line is pretty flat. Actually, it's a shrinker. It looks like it's been going downhill since the mid-60s, and it ends up in 2017 as low as it was in 1955. It's obviously not crowding anything out.

The blue line is way higher than the red because GDP is way smaller than All Sectors Debt. The blue line goes up because Household Debt is growing faster than GDP. The red line drifts downward because Household Debt is growing a little slower than All Sectors Debt.

What else does this graph show? It shows that if Household Debt is a problem because it is too high, then All Sectors Debt must also be a problem because it is too high. And that could be a useful thing to know.

// Part three tomorrow.

Thursday, February 8, 2018

The focus on Household Debt

It's almost like a ten-year anniversary: There is a lot of looking back on the 2008 crisis, a lot of pondering what might have been the core of the problem, and a lot of well-argued but incorrectly focused thought.

Happy anniversary.

In Rebuilding macroeconomics: Initial reflections on a major theory project, Konstantinos Efstathiou at Bruegel describes the view expressed by Paul Krugman:

For him, the failure to see the crisis coming does not pose a “deep conceptual issue” for economic models: it was not a lack of understanding of the possible mechanism” but a lack of attention to the right data” (e.g. overlooking institutional changes in the financial sector, the rise in household debt and looking at house-price growth in the aggregate rather than at a more local level).

The core of the problem, as Krugman sees it? The rise of household debt, the institutional changes that enabled it, and the rising price of houses that resulted from it. The problem for Krugman is household debt.

Krugman is not alone. A couple years back, Dirk Bezemer and Michael Hudson offered a "conceptual" argument:

Loans to non-financial business for production expand the economy’s investment and innovation, leading to GDP growth...

Like loans to non-financial business, household consumer credit provides the purchasing power and the effective demand for GDP to grow. But compared to business loans, it [leads to] less growth for the same loan amount, and more financial fragility.

In other words: Loans to consumers are not as good as loans to producers. It's the old "good debt vs bad debt" argument. The bad debt, as Bezemer and Hudson see it? Household debt.

They also studied "the growth effect of credit over time." They found the growth effect declining since the '80s, and "not significantly different from zero" since the '90s:

Credit was no longer “good for growth,” as many had for so long believed ...

A major reason for this trend was that credit was extended increasingly to households, not business.

My response: I hate arguments like that. Dangerous arguments. Arguments that make so much sense you think you don't need to look at the data.

Pushing the envelope, Bezemer and Hudson suggested that household borrowers crowded out borrowing by non-financial businesses:

While the total credit stock expanded enormously in the 1990s and 2000, credit to nonfinancial business was stagnant at about 40 percent of GDP, while its share in overall credit plummeted. By contrast, the share of household mortgage credit issued by banks rose from about 20 to 50 percent of all credit.

But it seems they lacked the evidence and the nerve to make that suggestion explicit.

In writing this post I came upon a remark yesterday that rang a bell in my head today. Of course, I can't find the article I was reading. But the guy said it was Mian and Sufi's book that focused everyone's attention on household debt.

Bezemer and Hudson are focused on household debt, for sure. But I'm not comfortable with their work. So after giving up the search for the article I read yesterday, I went searching for reviews of Mian and Sufi's book. This Google hit set me in the right direction:

"The government should have focused more on homeowner debt, and less on banks."

Well, yeah. But that's not what I read in Bezemer and Hudson. What I read there is that household debt is nonproductive and therefore "bad" and we shouldn't ever let it accumulate like that again. They're talking about what we can do to avoid this problem, next time. They are fighting the last war, the "excessive household debt" war.

What I read in the Google hit is entirely different. They're talking about what we should have done to fix the problem that now, ten years later, we hope we've recovered from. They're not trying to fight the last war. They are trying to understand how the economy works: The right way to recover from the crisis would have been to focus on reducing homeowner debt. Focus on the toxic liabilities of homeowners, rather than on the "toxic assets" of the banks. I went to the link.

Binyamin Appelbaum, 16 May 2014, NY Times: The Case Against the Bernanke-Obama Financial Rescue. Here is the opening:

Atif Mian and Amir Sufi are convinced that the Great Recession could have been just another ordinary, lowercase recession if the federal government had acted more aggressively to help homeowners by reducing mortgage debts.

Exactly: The money they used for QE -- they should have taken that money and used it to pay down debt for people. Pay down the toxic liabilities, make them non-toxic and make the toxic assets non-toxic at the same time. The money would have ended up in the same hands anyway: the hands of the creditors. And the liabilities that drag us down would have been lightened.

Isn't it obvious?

// A brief look at the data tomorrow.