Wednesday, March 14, 2012


At BBC News Business, Repressionomics - can 'financial repression' solve debt crisis? by Paul Mason:
It was economists Carmen Reinhart and Belen Sbrancia who, in March 2011, issued a ground breaking study of what "financial repression" means.

If we hear today the National Association of Pension Funds complaining that quantitative easing has placed a £90bn hole in the pension system, we can judge how rapidly the concept of "repression" is moving from theory to practice.

So what is "financial repression"? Put simply it is a combination of inflation and capital controls designed to erode the value of debts - and therefore of savings. It is overtly designed to prevent market mechanisms responding to inflation, leaving the price of borrowing too low and the return on savings too low.

Reinhart and Sbrancia pointed to the success of Western economies in "repressing" a mountain of debt after World War II - in a way that avoided fiscal austerity, and allowed a growth spurt, combined with inflation, to cancel out unsustainable debts.

No link, Paul?

The "mountain of debt after World War II" is not the same as today's mountain. After World War II it was a government mountain. The mountain of private debt that had created the Great Depression was gone, "eroded" by repayment, default, inflation, and the rocketing government debt of the second World War.

The mountain after that war was public debt; the mountain today is private debt.

And the "success" of Western economies in "repressing" debt after the war can be attributed largely to the economic growth that was made possible by the relative absence of private debt. But economic growth since the war was accompanied (and financed) by the growth of private debt.

Private debt grew until it hindered economic growth. Then government debt started growing again, and policies were put in place to encourage private credit use. But those policies failed to boost growth, because private debt was already excessive.

Reinhart and Sbrancia don't seem to see it that way:

Hoping that substantial public and private debt overhangs are resolved by growth
may be uplifting but it is not particularly practical from a policy standpoint. The
evidence, at any rate, is not particularly encouraging, as high levels of public debt appear to be associated with lower growth.

"High levels of public debt appear to be associated with lower growth." This is not true for the United States during the "golden" years after World War Two. High levels of public debt were associated with higher growth because private debt was low and did not interfere with private sector growth!

Related posts:
1. Debt Relatives
2. Debt Relatives: Uncle Sam
3. Debt Relatives: The Cousins
4. Debt Relatives: The Rise and Fall of the Non-Federal Relative


Anonymous said...

Good point. Everyone loves to point out the deleterious effects on growth of high public debt , always pointing to the R&R paper as proof , and leaving aside the bigger mountain of private debt.

This paper estimates similar growth-slowing effects for the other sectors ( household , corporate , etc. ) , and at similar thresholds ( ~80-90 debt/gdp )as R&R found for gov't debt :

"The real effects of debt"

So what happens when multiple sectors are above their growth-inhibiting thresholds at the same time? We're finding out right now.

We screwed up in the '70s when we let financialization run amok. If we had maintained limits of , say , 60-70% debt/gdp or debt/income for each of the household , corporate , and public sectors , with a hard limit of , say , 160% or so for the sum of the three ( a level that the post-war decades showed could support good growth and stability ) , we could have avoided this mess. A financial sector of the size we had during the '70s - running at a low debt/gdp of 10-20% - would have been more than sufficient to service such an economy.

Anonymous said...

Another example of someone who doesn't quite get it - McKinsey :

They think it's just peachy that household debt/DPI is about to reach its long-term trend line , shown in Exhibit 7 , at about 100% debt/DPI.

That long-term trend line is , of course , upward sloping , implying that sometime in the future they will consider a 10,000% household dept/DPI just peachy too.

The decline we want is one that approaches that flat spot in the '60s -'70s.

The Arthurian said...

Hello, Anonymous! Great comments. Great links. I think I know what I'll be reading (and writing about) this weekend.

I notice that your first link performs a thorough examination of non-financial debt -- probably what one would expect from the BIS.

Your analysis of the second link is beautiful... But I do like the graphs at that link.

Thanks for the links, the comments, and the visit.

Anonymous said...

Glad you found the links of interest.

Actually , many economists segregate financial from non-financial debts when analyzing economic growth , and I find it easier to think about the problem that way , too.

I think your characterization of financial/nonfinancial debt as productive/nonproductive hits on the reason I see it that way. Changes in private nonfinancial debt flows , at both the 1rst and 2nd derivative levels , correlate extremely well with private domestic economic activity (i.e. aggregate demand). Add in domestic gov't debt flows and gov't consumption and the relationship holds.

Now I'm not suggesting the financial sector debt is not a problem - I think it may be THE biggest problem - but I think it's a special case that should be looked at separately because of the way the finance sector exists as a parasite on the productive economy. The contribution of finance to overall GDP has only grown from ~2% 30 or 40 years ago to ~4% recently - almost a rounding error - while the finance sector debt has skyrocketed to levels that we can only guess at , given all the off-balance sheet trickery.

I'd like to see the recognition of the need to establish safe limits of debt/gdp for the various productive sectors ( based on studies like the one by Cecchetti et al ) , much as the EU attempted to do for sovereign debt in the Maastrict treaty , and then reign in the finance sector to a level that could service those demands , while hiving off the casino components into not-too-big-to-fail separate entities.

As you know , Keen includes financial sector debt in his "credit accelerator" models , but the group he got the idea from ("credit impulse" /Biggs et al ) focused on private nonfinancial debt:

See Fig 8 for the correlation of the credit impulse to private demand , and also take a look at Fig 10 , which shows their alternative take ( which strikes me as closer to the truth than conventional thinking ) on the output gap , showing the phony debt-fueled growth from 1996 to the crash.

I can dig up other links to their papers if you're interested.

Anonymous said...

Just one more quick thought on the "special case" of financial sector debt, inspired by a post I just saw on ZH :

This chart shows the changes in bank liabilities since the crash , looking at the shadow banking sector compared to the "traditional" banking sector :

The SB sector has delevered by about $6T since 2008 while traditional banks have added about $1T , for a net delevering of about $5 trillion.

Imagine if either the private sector or the gov't sector had delevered by an additional $5 trillion over that period , rather than the banking sector. The economy would be trashed , probably to a level comparable to the Depression. The reason this level of delevering of the SB system could occur so quickly with little notice is that much of that delevering is simply netting of offsetting positions between institutions.

There's a big problem in double-counting of financial sector debt that is either very temporary and offsetting - as in overnite repos - or is already counted on private sector balance sheets , as for mortgages and other consumer debt. That's why you see numbers all over the map for measures on the size of financial sector debt.

On the other hand , as I mentioned above , there may be a lot of financial sector debt that is invisible to standard accounting methods.

The Arthurian said...

The Biggs-Mayer paper reminds me very much of James Bullard's "earthquake" view that I found so interesting. And at 6 pages, it's not "too much" reading for me. I printed it out & expect to read it again today.

Jazzbumpa said...

Anon -

Perceptive comments. I think we are mostly on the same wavelength.

And you've increased my too-be-read list, for which I may have a hard time forgiving you.


Anonymous said...


Since you're interested in the Bullard debate , you might like to see this post by Felix Salmon , who "gets it" :

It amazes me that some of the brightest guys around - Krugman , Duy , Summer , etc. - dismissed the Bullard argument with a wave of the hand , arguing that since we didn't see inflation during the housing boom , and our arms haven't "fallen off" , then potential gdp hasn't changed , and we're still way below potential today.

The graphs used by Salmon for the U.S. are true on a global scale , too. We have unprecedented global overcapacity and malinvestment as a result of a decades-long global debt bomb , and to suggest that things will soon return to trend without major structural changes - also on a global scale - is ludicrous. The economics profession will be embarrassed by events once again , but they must be getting used to it , I suppose.

We could return to trend growth , both in the U.S. and globally , with the right structural changes - like a 180-degree turn in the direction of wealth and income redistribution , from upward ( currently ) to downward ( as in the "Golden Age" decades ). A massive debt jubilee would be a good way to start , as would high top marginal rates and a return of confiscatory estate taxes on the super-wealthy.

While the arms of the masses still have their productive capacity , they've been drained of their consumption capacity. Until that capacity is restored , potential gdp revisions will go in only one direction - downward.