It looked on Wednesday last week like Felix Salmon had had the last word on what he earlier dubbed the Kanjorski meme – a little piece of web flotsam alighted upon by a number of blogs, among them FT Alphaville – the gist of which went something like this:
Within 24 hours the world economy would have collapsed.
On Thursday (Sept 18), at 11am the Federal Reserve noticed a tremendous draw-down of money market accounts in the U.S., to the tune of $550 billion was being drawn out in the matter of an hour or two. The Treasury opened up its window to help and pumped a $105 billion in the system and quickly realized that they could not stem the tide.
Felix, sceptical from the off, appeared to have put things to bed:
…there never was a $500 billion outflow from any asset class in the space of a couple of hours or even weeks, and the Fed never shut down or froze any money-market accounts.
In fact, writes Salmon, notwithstanding the dramatic withdrawal requests from the Reserve Primary fund (which broke the buck on September 15, when Lehman failed), money market funds, though roiled, were not completely collapsing.
The news from The Reserve was gruesome, and total withdrawals from money-market funds reached $104 billion that day, according to Crane Data. Another data provider, ICI, says that as of the close of business on the 17th, money-market funds had a total of $3,549.3 billion, which was a fall of just $30.3 billion from their level a week previously.
The following day, September 18, was bad but not quite as bad, with withdrawals of $57 billion, according to Crane Data. By the 24th, according to ICI, the total was $3,456.2 billion — a drop of another $93.1 billion from the 17th.
Now firstly, there’s a problem with looking at the MM fund market as a whole. There are three types of MM fund – those that invest in corporate commercial paper, those that invest in US Treasuries and those that invest in other government bonds. The really dramatic problem in the money markets – the one which, as Kanjorski intones was tantamount to “an electronic run on the banks” – was the shift within the money market fund universe, specifically, the massive redemptions from bank commercial paper-investing funds (called “prime funds) and an almost consummate increase in deposits at Treasury and Government funds. It’s nicely illustrated by this Bank of America graph, which like Felix, uses Crane data:
Looking at the fall in size of the money market fund universe in aggregate is something of a canard. It certainly doesn’t show the crisis quite for what it was – a total collapse in prime funds – and with that, an acute liquidity crisis for any corporate institution with a sizeable CP facility.
Secondly, there’s the figure at the heart of the Kanjorski meme: the $550bn of withdrawals from money market funds on Thursday September 18th. Salmon suggests that the number originated in no less reputable a place than the New York Post, which on September 21 wrote:
According to traders, who spoke on the condition of anonymity, money market funds were inundated with $500 billion in sell orders prior to the opening [on Thursday]. The total money-market capitalization was roughly $4 trillion that morning.
But David Merkel at the Aleph blog may, in fact, have something which corroborates the provenance of the Kanjorski meme; and most notably, that some of the numbers in it came from Hank Paulson. The below is an extract from a research report (authored by Congressman Jim Saxton) to the Joint Economic Committee of Congress (emphasis ours):
Irrational runs on money market mutual funds began. For the week ending on Wednesday September 17, 2008, investors redeemed $145 billion from their money market mutual funds. On Thursday September 18, 2008, institutional money managers sought to redeem another $500 billion, but Secretary Paulson intervened directly with these managers to dissuade them from demanding redemptions. Nevertheless, investors still redeemed another $105 billion. If the federal government were not to act decisively to check this incipient panic, the results for the entire U.S. economy would be disastrous.
In other words, institutional clients tried to pull around $500bn, but were dissuaded by the Treasury Secretary – who must then have also been hectically working on the money market fund insurance programme, announced only the following day.
Such a subtle correction to the Kanjorski meme answers a lot of questions.There were requests for $500bn of redemptions, but not actually $500bn of redemptions. And it feels right too. FT Alphaville is aware of very similar circumstances back in September 2007 when secretary Paulson rang around various money market funds to dissuade them themselves from pulling money from a number of ailing bank SIVs (which were dependent on CP for daily financing). Rating agencies got similar calls.
And, anyway, zooming out slightly, is $500bn really such a big number in context? Reserve Primary breaking the buck was a phase transition – it completely altered the market and the psychology of it. Put yourself in the position of a huge institution with billions stashed in a money market fund – the equivalent of a personal bank account, as far as such institutions are concerned – you have no insurance and there’s a very very significant risk you’ll lose money if you keep it where it is while everyone else is redeeming. It’s a bit of a no-brainer. Considering prime funds had around $1.9 trillion in them before Lehman’s collapse, $500bn isn’t that much.
Now whether you follow through with Kanjorski on the conclusion that “within 24 hours the world economy would have collapsed,” and that “it would have been the end of our economic system and our political system as we know it,” is another matter.