The Fed wants banks to be able to survive epic Fed incompetence

A commenter named Steve sent me this puzzling article on stress tests:

The Federal Reserve plans to stress test six large U.S. banks against a hypothetical market shock, including a deterioration of the European debt crisis, as part of an annual review of bank health.

The Fed said it will publish next year the results of the tests for six banks that have large trading operations: Bank of America (BAC.N), Citigroup (C.N), Goldman Sachs (GS.N), JPMorgan Chase (JPM.N), Morgan Stanley (MS.N) and Wells Fargo (WFC.N).

“They are clearly worried about the issue of Europe,” said Nancy Bush, a longtime bank analyst and contributing editor at SNL Financial. “In a time of risk aversion and concern, you need transparency.”

The Fed said its global market shock test for those banks will be generally based on price and rate movements that occurred in the second half of 2008, and also on “potential sharp market price movements in European sovereign and financial sectors.”

In the Fed’s hypothetical stress scenario, unemployment would spike as high as 13 percent while U.S. gross domestic product would fall by as much as 8 percent.

I’d rather the Fed stress test their own policies.  Can they keep US NGDP rising at a 5% rate as Europe goes to pieces?  If not, what sort of policy regime might be able to do that?



16 Responses to “The Fed wants banks to be able to survive epic Fed incompetence”

  1. Gravatar of anonymous anonymous
    24. November 2011 at 07:57

    How about the Bank of England?

    In the UK two year inflation (RPI) break evens have fallen from 2.90% to 2.34% in the two months after the MPC announced further asset purchases in the October meeting.

    Yields are now negative for the entire inflation indexed government curve.

    The minutes from the most recent MPC meeting in November state:

    “The Committee noted that the existing programme of asset purchases would take a further three months to complete and market capacity made it difficult to increase the monthly rate of purchases substantially above what was already under way.”

    I.e. in the short term they do not believe they can further expand asset purchases.

    Despite the asset purchases, short term inflation expectations are plummeting as they did after immediately before Lehman failed in September 08; inflation expectations (breakevens) did not recover until mid January 2010.

    What should the MPC do when they meet in two weeks time? Would a nominal GDP target be sufficient?

    The BoE has already purchased over 25% of outstanding government debt, how much government debt would they have to purchase to meet the target 50%? 75%? 100%?

    What happens if they purchase all government debt and inflation expectations have still not recovered?

    I suppose they could make a start on buying, for GBP, all Euro denominated debt from UK based financial institutions. To be honest that probably won’t cost that much by next week.



  2. Gravatar of Peter N Peter N
    24. November 2011 at 10:24


    “I suppose they could make a start on buying, for GBP, all Euro denominated debt from UK based financial institutions. To be honest that probably won’t cost that much by next week.”

    England is a major financial center, so the ratio of bank assets to GDB is probably too large for this. Now if the ECB did it for everybody and issued Eurobonds, that might work. If countries wanted to issue more debt, they’d have to get the ECB to agree to issue more Eurobonds, since nobody would buy individual country bonds that were subordinate to their debt to the ECB – at least not at an affordable price.

    Maybe Germany would be OK with discipline enforced through the ECB. You could include the IMF in the deal and get more political cover and experience with enforcement.

    You might be able to do this without a treaty change, which would be a nightmare of logrolling.

  3. Gravatar of Peter N Peter N
    24. November 2011 at 10:44


    I think the Fed wants to test the banks’ ability to handle the ECB’s exaggerated imitation of the Fed’s 2008 incompetence.

    There’s a limit to what the Fed can do post Dodd-Franks (just a bit short sighted, wasn’t it).

    If you liquidate all the major European banks, the damage this does to their US counterparties will be very hard to deal with through conventional monetary policy. It’s not well targeted enough. Expectations don’t trump insolvency.

    The Fed will need all the tricks from 2008 and a few new ones. We need J.P. Morgan locking the bankers in his library until they agreed on a solution and funded it.

  4. Gravatar of Steve Steve
    24. November 2011 at 15:23


    Thanks for doing a post on this. I think it’s a really important topic as it has the potential to become self-fulfilling. Stress-testing involves a delicate dance between being strong enough to be credible and weak enough not to be self-fulfilling. Often times, the stress test is really just a stand in for some other policy certainty that market participants need.

    1) Mar ’09 US stress tests banks — some people think it is too easy, but what it does is provide clarity that there will be no nationalization. Markets rally, not because the stress tests were easy or successful, but rather because they provided certainty of capital needs.
    2) Early ’11 EU stress tests banks — no haircuts on sovereigns. Test doesn’t have credibility because no one believes Germany will stand behind sovereigns. Markets tank, not because the stress tests were too easy, but because the need for fiscal union and/or monetary ease were ignored.

    Based on 1 and 2, people (wrongly) conclude stress tests need to be harsh in order to succeed. So we get:

    3) EU to apply mark-to-market accounting on sovereign debt for stress tests. Well, German debt trades aroung 105% of par, while Italian debt is closer to 90% of par. Therefore, Italian banks must liquidate holdings in a feedback loop, while presumably Germans can buy up distressed assets. It’s almost as if the accounting rules were written by Germans intent on waging economic warfare on Italy. Anyway, the stress test fails even though it applies MTM accounting because it doesn’t address the fundamental issues of fiscal union, ECB debt buying, and the possibility of debt haircuts.
    4) US announces -8% RGDP +13% jobless stress test. The risk here is that it is self-fulfilling because now banks have to price loans against their cost of capital subject to this armageddon scenario. Bank lending should be hit as banks seek to further shrink and derisk their balance sheets. To what end? Has the Fed provided any policy certainty here? NO! So there’s no mileage to be gained by testing for it. Much more could be gained by a commitment to reflate back to trend if such a scenario unfolded. A reflation committment also carries with it the potential to be self-fulfilling by mitigating such a dire outcome in the first place.

  5. Gravatar of James in London James in London
    25. November 2011 at 02:51

    a great way of putting it. but you still need to make it clear how you would reform the banksters. all evidence points to the big wholesale banks in the list as having unsustainable business models without the implicit state guarantee. for instance, would you support a new Glass-Steagall?

  6. Gravatar of ssumner ssumner
    25. November 2011 at 07:16

    anonymous, They have lots of options, but the most important is deciding where they want to go–set a higher NGDP target.

    If they do so, they’d probably have to sell assets. But if not they can cut interest rates, or lower IOR, or buy US T-securities, devalue the pound, the options are endless.

    But the key is setting a target, it’s the sine qua non of policy.

    Peter, You said;

    “There’s a limit to what the Fed can do post Dodd-Franks (just a bit short sighted, wasn’t it).”

    I’m no fan of Dodd-Frank, but I would hope it limited the Fed’s ability to do bank bailouts. Then their only option would be to maintain steady growth in NGDP.

    Steve, Very good observations.

    James, NGDP targeting is the necessary fist step in bank reform. That removes the fear of bank failures driving the economy into recession.

    The next step is removing too big to fail. Then we need to reform FDIC (I’m assuming it will be politically impossible to end it.) Ban sub-prime mortgages with FDIC-insured money. Also ban loans for commercial real estate projects with FDIC-insured funds. Force banks to raise funds used on those sorts of risky projects via other methods. Use FDIC-insured money solely for investments in government bonds and 20% plus downpayment mortgages. These reforms don’t end the moral hazard problem, but they greatly reduce it.

  7. Gravatar of dwb dwb
    25. November 2011 at 07:16

    lol.yeah, the fed should stress test its own policies. excellent idea. ECB too.

  8. Gravatar of Wicked Lad Wicked Lad
    25. November 2011 at 08:00

    Regarding the “epic Fed incompetence,” what can the Fed do to manage NGDP? I see how they can influence inflation and interest rates, but I don’t see how they can do anything significant about unemployment (one of their current mandates) or NGDP (your recommendation). So I don’t see how it’s epic Fed incompetence if GDP falls.

    For a little context, I’m an utter economics amateur and a regular Krugman reader expanding my economics breadth.

  9. Gravatar of Steve Steve
    25. November 2011 at 09:02

    ssumner wrote: >>NGDP targeting is the necessary fist step in bank reform. That removes the fear of bank failures driving the economy into recession.<<

    I would argue that NGDP targeting removes the fear of recession driving the banks into failure. (at least all at once)

  10. Gravatar of James in London James in London
    26. November 2011 at 01:20

    Abolishing the FDIC is not a bad idea, but you should recognise that they have been the ONLY regulatory body attempting to protect taxpayers during the bank failures: taking the banks over, sacking the management, breaking them up and selling the parts. A near-market solution. All other agencies seem to just want to bail them out and leave the banksters in charge, and free to bribe (sorry, lobby) the politicians to let them alone to do what they want all in the TBTF implied guarantee comfort zone.

    Starting with the FDIC is a bit harsh. But then you are now very closely aligned to those arch-banksters at Goldman Sachs with their self-avowed strategy of making themselves Too Big, Too Crucial To Fail. And they have the ex-personnel in all key areas of US and European policy to guarantee that. You must be careful who you choose for your friends.

  11. Gravatar of ssumner ssumner
    26. November 2011 at 07:18

    Wicked lad, You concede they can control inflation. Well if inflation rises then NGDP rises unless RGDP falls. But the policy that would boost inflation (expansionary monetary policy) does not cause RGDP to fall, indeed it might cause it to increase. So if you are right that the Fed can influence inflation, then ipso facto they can influence NGDP.

    Steve, That too.

    James, G-S are not my friends. Bloggers don’t have friends, they have interests.

    Our economy has lots of big problems, G-S isn’t one of them.

    FDIC has required a more than $100 billion taxpayer bailout. That’s a disastrous performance. Don’t forget that the banks they are closing engaged in risky practices precisely because FDIC insurance made it deposits safe, so that depositors didn’t care, and hence no longer monitored their activity.

  12. Gravatar of James in London James in London
    26. November 2011 at 13:04

    I agree depositors should have no insurance from the state. I agree they should be made to care by occasionally losing some or all of their deposits. It will be good for society in the long run.

    GS are just superbly aware of the arbitrage of implicit state guarantees, and indeed have built their business model on making sure they have so many fingers in so many pies that they are way to important to fail. Mostly rather obscure but systemic parts of the financial markets. They’ve also enjoyed bailouts directly, and indirectly by the bailout of their major counterparties (remember AIG and what Maiden Lane was all about).

    While GS keep to this business model there will be no end to this type of behavior as so many banks see the “success” of GS and want to copy it. JPM is little better. Super-size Citi and BoA were built on the same principle and currently suffering the diseconomies of scale that eventually throws them onto state (taxpayer) charity.

    If you don’t “get it” about this arbitrage you have no chance of working at GS or one of its super-size peers. Luckily you don’t, so hopefully we won’t ever see you working for the “vampire squid”.

  13. Gravatar of Wicked Lad Wicked Lad
    26. November 2011 at 14:25

    Scott Sumner wrote, “So if you are right that the Fed can influence inflation, then ipso facto they can influence NGDP.”

    Then I don’t get why an NGDP target is better than an inflation target. But I found your Defense of NGDP Targeting essay and will read that. I should have done that first, I suppose. Thank you, Dr. Sumner.

  14. Gravatar of Scott Sumner Scott Sumner
    27. November 2011 at 12:28

    James, I don’t know how much of G-S’s success is due to rent seeking, so I won’t argue the point.

    And I never had any expectation of working there. I’m thinking of asking Corzine for a job.

    Wicked Lad, Yes, read that essay.

  15. Gravatar of James in London James in London
    28. November 2011 at 02:37

    G-S, like the very best rent seekers, try their damnedest to keep it well hidden.

    Asking Corzine for a job, are you going to jail too?

  16. Gravatar of Scott Sumner Scott Sumner
    28. November 2011 at 17:42

    I was joking about Corzine, but I suppose you were too.

Leave a Reply