The Fed’s $29 Trillion Bailout of Wall Street
UPDATE: to read the Working Paper (“$29,000,000,000,000: A Detailed Look at the Fed’s Bailout by Funding Facility and Recipient”) click here
Since the global financial crisis began in 2007, Chairman Bernanke has striven to save Wall Street’s biggest banks while concealing his actions from Congress by a thick veil of secrecy. It literally took an act of Congress plus a Freedom of Information Act lawsuit by Bloomberg to get him to finally release much of the information surrounding the Fed’s actions. Since that release, there have been several reports that tallied up the Fed’s largess. Most recently, Bloomberg provided an in-depth analysis of Fed lending to the biggest banks, reporting a sum of $7.77 trillion. On December 8, Bernanke struck back with a highly misleading and factually incorrect memo countering Bloomberg’s report. Bloomberg has—to my mind—completely vindicated its analysis; see here.
Any fair-minded reader would conclude that Bernanke’s memo to Senators Johnson and Shelby and Representatives Bachus and Frank is misleading. One could even conclude that it is not just a veil of secrecy, but rather a fog of deceit that the Fed is trying to throw over Congress.
He argues that the sum total of the Fed’s lending was a mere $1.2 trillion, and that it was spread across financial and nonfinancial institutions of all sizes. Further, he asserts that the Fed never tried to hide the bail-outs from Congress. Both of these assertions fly in the face of the facts available (as the Bloomberg response makes clear).
As Bernanke notes, highly credible analyses of the bailout variously put the total at $7.77 trillion (Bloomberg) to $16 trillion (GAO) or even $24 trillion (I think this is Senator Bernie Sanders’ figure). He argues that these reports make “egregious errors,” in particular because they sum lending over time. He also claims that these high figures likely include Fed facilities that were never utilized. Finally, he asserts that the Fed’s bailout bears no relation to government spending, such as that undertaken by Treasury.
All of these assertions are at best misleading. If he really believes the last claim, then he apparently does not understand the true risks to which he exposed the Treasury as the Fed made the commitments.
There are a number of issues that must be understood. First, the Fed quibbles about the differences among lending, guarantees, and spending. For the purposes of this blog I will accept these differences and call the sum across the three “commitments.” In spite of what Bernanke claims, these do commit “Uncle Sam” since Fed losses will be absorbed by the Treasury. (The Fed pays profits to Treasury, so if its profits are hurt by losses, payments to Treasury are reduced. If the Fed should go insolvent, the Treasury will almost certainly be forced to recapitalize it.) I do, however, agree with the Chairman that a tally should not include facilities that were created but not utilized (there were several of them, and the tally I present below does not include any facilities that were not used).
Second, there are (at least) three different ways to measure the Fed’s bailout. One way would be to find the day on which the maximum outstanding Fed commitments was reached. According to the Fed, that appears to have been about $1.5 trillion sometime in December 2008. I’m willing to take Bernanke at his word. Another way would be to take the total of commitments made over a short period of time—say, a week or a month. That would be a measure of systemic distress and would help to identify the worst periods of the GFC (global financial crisis). Obviously, this will be a bigger number and will depend on the rate of turnover of Fed loans. For example, many of the loans were very short-term but were renewed. Bernanke argues that it is misleading to add up across revolving loans. Let us say that a bank borrows $1 million over night each day for a week. The total would be $7 million for the week. In a period of particular distress, the peak weekly or monthly lending would spike as many institutions would be forced to continually borrow from the Fed. Bernanke argues we should look only at the lending at a peak instant of time.
Think about it this way. A half dozen drunken sailors are at the bar, and the bartender refills their shot glasses with whiskey each time a drink is taken. At any instant, the bar-keep has committed only six ounces of booze. That is a useful measure of whiskey outstanding. But it is not useful for telling us how much the drunks drank. Bernanke would like us to believe that if the Fed newly lent a trillion bucks every day for 3 years to all our drunken bankers that we should total that as only a trillion greenbacks committed. Yes, that provides some useful information but it does not really measure the necessary intervention by the Fed into financial markets to save Wall Street.
And that leads to the final way to measure the Fed’s commitments to propping up our drunks on Wall Street: add up every single damned loan, guarantee and asset purchase the Fed made to benefit banks, banksters, real Housewives on Wall Street, fraudsters, and their cousins, aunts and uncles. This gives us the cumulative Fed commitments.
The final important consideration is to separate “normal” Fed actions from the “extraordinary” or “emergency” interventions undertaken because of the crisis. That is easier than it sounds. After the crisis began, the Fed created a large alphabet soup of special facilities designed to deal with the crisis. We can thus take each facility and calculate the three measures of the Fed’s commitments for each, then sum up for all the special facilities.
And that is precisely what Nicola Matthews and James Felkerson have done. They are PhD students at the University of Missouri-Kansas City, working on a Ford Foundation grant under my direction, titled “A Research And Policy Dialogue Project On Improving Governance Of The Government Safety Net In Financial Crisis.” To my knowledge it is the most complete and accurate accounting of the Fed’s bailout. Their results will be reported in a series of Working Papers at the Levy Economics Institute. The first one is titled $29,000,000,000,000: A Detailed Look at the Fed’s Bailout by Funding Facility and Recipient. Read it here.
Here’s the shocker. The Fed’s bailout was not $1.2 trillion, $7.77 trillion, $16 trillion, or even $24 trillion. It was $29 trillion. That is, of course, the cumulative total. But even the peak outstanding numbers are bigger than previously reported. I do not want to take any of their fire away—interested readers must read the full account. However, I will use their study as the source for a brief summary of total Fed commitments.
Here I am only going to focus on the final measure of the size of the bailout: the cumulative total. This is not directly comparable to the Fed’s $1.2 trillion estimate, which is peak lending. It is closest to the $24 trillion figure that I believe Senator Sanders is using. The difference from that number is probably attributable to choice of facilities to include.
I will post more on the important research done as part of this Ford Foundation grant; in coming blogs I will also explain why all Americans should be horrified at the Fed’s actions, and by Bernanke’s continued attempt to cover-up what the Fed has done.
Summary of Total Cumulative Fed Commitments
When all individual transactions are summed across all facilities created to deal with the crisis, the Fed committed a total of $29,616.4 billion dollars. This includes direct lending plus asset purchases. Table 1 depicts the cumulative amounts for all facilities; any amount outstanding as of November 10, 2011 is in parentheses below the total in Table 1. Three facilities—CBLS, PDCF, and TAF—overshadow all other facilities, and make up 71.1 percent ($22,826.8 billion) of all assistance.
Table 1: Cumulative facility totals, in billions
Source: Federal Reserve
| Facility | Total | Percent of total |
| Term Auction Facility | $3,818.41 | 12.89% |
| Central Bank Liquidity Swaps | 10,057.4(1.96) | 33.96 |
| Single Tranche Open Market Operation | 855 | 2.89 |
| Terms Securities Lending Facility and Term Options Program | 2,005.7 | 6.77 |
| Bear Stearns Bridge Loan | 12.9 | 0.04 |
| Maiden Lane I | 28.82(12.98) | 0.10 |
| Primary Dealer Credit Facility | 8,950.99 | 30.22 |
| Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity Facility | 217.45 | 0.73 |
| Commercial Paper Funding Facility | 737.07 | 2.49 |
| Term Asset-Backed Securities Loan Facility | 71.09(.794) | 0.24 |
| Agency Mortgage-Backed Security Purchase Program | 1,850.14(849.26) | 6.25 |
| AIG Revolving Credit Facility | 140.316 | 0.47 |
| AIG Securities Borrowing Facility | 802.316 | 2.71 |
| Maiden Lane II | 19.5(9.33) | 0.07 |
| Maiden Lane III | 24.3(18.15) | 0.08 |
| AIA/ ALICO | 25 | 0.08 |
| Totals | $29,616.4 | 100.0% |
Source: “$29,000,000,000,000: A Detailed Look at the Fed’s Bailout by Funding Facility and Recipient” by James Felkerson, forthcoming, Levy Economics Institute, based on data analysis conducted with Nicola Matthews for the Ford Foundation project “A Research And Policy Dialogue Project On Improving Governance Of The Government Safety Net In Financial Crisis.”
ShareThis
There are a lot of very important criticisms to be made of the fed and the banks, but there is a lot of, ah, let’s say under-informed analysis getting put out there, too.
I’d be really curious to learn where you learned your banking stuff from. I didn’t see anything especially banking related on your CV.
“There are a number of issues that must be understood. First, the Fed quibbles about the differences among lending, guarantees, and spending. For the purposes of this blog I will accept these differences and call the sum across the three “commitments.”
You really don’t understand the difference between lending and guarantees? You think that is quibbling? And you think they can be lumped under the term “commitments” for the purpose of analyzing the many important finance issues facing our economy?
Maybe things have changed big time since I practiced banking law (corporate and int’l lending, trade finance and reg comp, mainly) from the 70s to 90s for a money center and then super-regional bank, but I suspect not when it comes to the basics.
For the purpose of analyzing banking and the Fed’s support thereof,do you understand that there are significant differences between secured lending and unsecured lending? Between those sorts of lending and commercial paper facilities? And between those and currency swap facilities? Just for example.
The drunken sailor analogy is delightfully graphic but it misses a vital point – on a revolving facility, where there are multiple, sequencial drawdowns, the first drawdown (let’s assume it’s for the full amount of the line) has to be paid back before the second drawings are permitted. That payback feature is entirely missing from the drunken sailor analogy. (And it is likely, but I don’t know for certain, that even if a facility wasn’t fully drawn, there would have been conditions precedent such as paying down the balance in accordance with a repayment schedule to subsequent draw down.)
(I very much like the drunken sailor analogy as it generally applies to “Wall Street,” – it really captures the outrageous excesses of Wall Street over the last 15- 20 years or so – but it just doesn’t work the way it is being applied here.)
Trying to lump all drawings over the course of such a facility and portray that lump sum as the significant figure doesn’t make a lot of sense, at least from a risk analysis or even an accounting point of view.
And let’s look at the different types of lending under the fed programs program.
Yes, some of it seems to have unusually risky and manipulative.
But how about the commercial paper lines? One big criticism of the TARP and other support programs is that the money went to Wall Street, not Main Street.
A commercial paper facility supports day to day commercial financing for “Main Street.’ Commercial paper (typical 1 day to 270 days) is used for day to day commerce – funding inventory, payrolls and other commercial operations. (What would be interesting to know is what was the CPFF discount rate charged to the banks and what rates were they changing to the borrowers, but that isn’t the issue here.)
BTW commercial paper is about as non-risky as any form of bank lending (Look at how it is treated under the Risk Based Capital Adequacy Quidelines and the Basel Accords.) At that point in time, it seems to me, the bigger the commercial paper facilities, the better our economy was served.
And I don’t know where the “$10” trillion in currency swaps show up in your chart, but they were swaps, not loans and the amount at risk in a currency swap is well below the nominal amount of the facility, especially as as one currency is held as collateral for the currency distributed.
[BTW, did you make the same mistake as Alan Grayson did and think the Swiss Bank Currency Swap line was for $60 billion? Or did you understand that is was for the purchase at a discount of $60b in face value assets – apparently purchased by the SPV for $13B. (Yeah, the Fed report on that detail is a little bit ambiguous. Publication page 129, .pdf page 142.)]
The secured nature of the currency swaps is another reason why trying to lump everything under the term “commitments” is misleading, as is the attempt to measure exposure by simply adding all tranches, loans, swaps, guarantees, together.
I am as liberal as they come, and I believe there are hugely significant issues regarding and criticisms of banking and the Fed, especially after almost 30 years of being lead by a libertarian Ayn Rand acolyte.
But if we’re going to fix these problems, we have have at least some basic knowledge about what was actually going on and about how various banking and finance products work.
Somewhat parallel comments are at http://rjw-progressive.blogspot.com/2011/12/needless-fed-freakout.html
I’m a huge admirer of your writings, so it dismays me to read this, which really seems (to me) to utterly misrepresent the situation. James Hamilton’s arguments seem irrefutable to me.
Again, this from a huge fan of yours who is strongly predisposed to take everything you write as gospel.
OK, Steve Roth and others: look, my blogs as well as James Felkerson’s paper make it clear that there are three different ways of measuring the size of the Fed’s bail-out. I won’t repeat in detail here how each is done, and why each is useful, since anyone who is interested can take the time to read our pieces. The favored one used by the Fed and its supporters and apologists is the smallest one: look for the day on which outstanding loans reached the peak–about $1.2 trillion. They then argue that no other measure is appropriate. I think that is quite silly. Let me demonstrate by using the critique made by James Hamilton–who wants to minimize the size of the bail-out. That is not too surprising since he proudly acknowledges that he’s been working for the Fed for twenty years.
Here is his critique of our work:
“But Professor Wray goes on to speak admirably about an analysis by his student James Felkerson that does exactly that, and concludes that the Fed lent not $7.77 trillion but instead $29 trillion. For example, Felkerson takes the gross new lending under the Term Auction Facility each week from 2007 to 2010 and adds these numbers together to arrive at a cumulative total that comes to $3.8 trillion. To make the number sound big, of course you want to count only the money going out and pay no attention to the rate at which it is coming back in. If instead you were to take the net new lending under the TAF each week over this period– that is, subtract each week’s loan repayment from that week’s new loan issue– and add those net loan amounts together across all weeks, you would arrive at a cumulative total that equals exactly zero. The number is zero because every loan was repaid, and there are no loans currently outstanding under this program. But zero isn’t quite as fun a number with which to try to rouse the rabble.”
Ok, by Hamilton’s logic, the Fed’s bail-out was zero. Wall Street solved its own crisis. Since, after all is said and done, TAF cumulative net lending is zero. Pretty silly.
What really happened is that the Fed had to keep lending month after month because whenever a short-term loan was repaid, the banks could not fund their positions by turning to markets. They had to go back to the Fed and ask for more help. Markets did not trust them, because the banks were troubled. Here we are now, almost 4 years after the crisis began and the biggest banks are still troubled.
Our argument is that over the period of three years analyzed, the Fed had to make continued massive interventions–indeed, an effort that is entirely unprecedented–to save Wall Street. Three years after the crisis, these megabanks still could not fund themselves in markets. They had to rely on the Fed’s assistance. We never have claimed that the Fed’s risk exposure was $29 trillion. Of course not. We are getting a measure of the Fed’s continued extension of life support to unhealthy banks.
Hamilton’s favored measure–outstanding loans in December 2008–sheds very little light on the fact that the Fed had to continually pump new lending for years to save Wall Street. He is setting up a strawman argument that we never made. We are well aware that the cumulative measure is not a measure of the Fed’s risk exposure at any point in time.
With regard to all of his criticism of the Bloomberg study, I do not have any comment–except that our study DID NOT include any facilities that were created but not used. He claims that Bloomberg did. I do not know if he is correct. But he is wrong to argue that ex ante guarantees made by the Fed do not entail risk merely because ex post they were not used. If I promise to pay all damages caused by a drunk when I let him drive my car home, I am at risk. If he arrives home safely, I cannot claim there was no risk in the guarantee.
[…] for example this item from the Levy Institute blog written by University of Missouri Professor L. Randall Wray, which […]
Now that’s more like the kind of commentary I expect from Randall Wray. “Here are different ways to look at this, and what we learn from each.”
Yes, Hamilton’s ultimate (if implicit?) *message* — the financial markets solved everything! — has to be taken down. But this current post as framed and presented, I think, gives aid and comfort to the enemy, and hurts your reputation as a voice of sober reason. Hamilton’s post is #1 on Reddit’s economics feed right now, and if others are like me, they’re giving it a lot of credence cause it makes sense. But they don’t know your body or work, so…they might not, ever.
Also, if others are like me, they’re really wondering not so much how much the banks borrowed (since it does seem to be getting paid back, mostly…), as how much their bottom lines ended up being subsidized, explicitly and implicitly. (Maybe you’ve addressed this and I’ve missed it.) The confusion out there on that topic (including mine) is profound. I just have no idea, even the order of magnitude.
Would love to see more work like Haldane’s, with such work widely publicized and discussed:
http://www.voxeu.org/index.php?q=node/7314
I am a regular reader of econbrowser and was linked here from that site.
Fed is the lender of last resort in our fractional reserve banking system. Surely the Fed can loan money to banks facing a run? These temporary loans were paid back in full with interest.
It is simply the Fed functioning as designed.
During the 1930’s the Fed did not have the ability to “lend money into existence”. As Milton Friedman correctly it was the Fed’s inability (or unwillingness) to lend to commercial banks which caused bank runs and turned a bad recession into the great depression.
[…] for example this item from the Levy Institute blog written by University of Missouri Professor L. Randall Wray, which […]
I agree with rjw.progressive that your drunken sailor analogy just doesn’t hold water. It doesn’t make sense – a loan is not like taking a drink because you (should) pay the former back.
If the Fed had literally given money to the banks and hadn’t wanted repayment, then your approach is right – sum up all of the largesse because that’s the handout from the Fed to the bankers. And there, the drunken sailor analogy works fine.
In the context of these lending operations, the Fed had exposure to losses. But that exposure wasn’t cumulative. Instead, as a loan was repaid, that risk was removed. Now, the loan might be rolled over. Fine. But that doesn’t change the Fed’s exposure to losses. It gave out a 100 loan, faced that potential loss, got the money back, loaned it out again, and still faced the potential loss of 100. You may say, “But the Fed could have lost 200!” That doesn’t make any sense. Assuming that the Fed wants to get repaid, their exposure is conditioned on the loans that they grant, and those loans are conditioned on the repayment of prior loans. Unless the Fed honestly doesn’t care about getting repaid in which case your calculation is exactly correct. Otherwise, if you’re interested in how much support the Fed provided, you should look at how much the Fed could have lost, and I think that the maximum is then the appropriate metric.
Here’s what Randy sent me in response to my post:
“Good for you
I guess you also do not really follow the argument. It is not about exposure to risk.”
Here’s what I sent in response:
It’s not about exposure to risk, you say. Umm, ok, so what is the issue with the bailout? If it involved a bunch of money passed around with no threat of any losses, then who cares? In that world, everything gets repaid, so we can sum up the amount lent as much as we want, but it doesn’t reflect anything meaningful. The entire reason that the bailout mattered is b/c the taxpayer could have been on the hook for losses that otherwise would’ve been incurred by bondholders, shareholders, etc. So I think that the first order concern is, in fact, the amount of money that was at risk. And, as was noted earlier in comments on (and subsequent articles about) your post, the money at risk is not cumulative. You cannot add the total amount lent and claim that this somehow reflects the amount of exposure faced by the taxpayer due to the bailouts.
Thanks again for your response. Sorry if you don’t understand how lending works.
The drunken sailor analogy is quite apt. But it needs some amplification provided by the wisdom of the ages, past & future.
As a wise man said – probably Sumerian, they invented & said everything – you don’t buy beer, you rent it. Where are the drunken sailors? A unniversal truth can be easiest to see at Quark’s bar on Deep Space 9. They provide liquidity the old-fashioned way. They recycle it.
Wall Street really did get a $29T buzz, have a $29T party.