Tuesday, August 19, 2008

Watching the Credit Crisis Like the Pros

Credit crisis... credit crunch... we all hear the terms, but what do they mean to the average schmoe?  Well, for one thing, it means that even though the Fed has lowered the 'Fed Funds' rate to 2%, and the yield on 10-year US Treasuries is around 3.80%, rates on things like mortgages, car loans, and credit cards are persistently high.  For example, Freddie Mac - one of the two big agencies that guarantees mortgages, the other being Fannie Mae - says that 30-year conventional, conforming mortgages averaged 6.52% with 0.7 points last week.  Conventional and conforming means that the borrower and the loan size meet Freddie's ever-stricter requirements, and points, for those who've never had a mortgage, are upfront costs.  One point is one percent of the loan - so the more points, the more expensive it is to get the loan.  Contrast this with the situation at the beginning of this year, when the credit crisis was already in full swing - 10-year yields were 0.20% lower, at 3.60%, but a 30-year mortgage guaranteed by Freddie was 0.45% lower, at 6.07% with only 0.5 points.  Go back further and the difference is even more dramatic - in the first week of January 2006, 10-year Treasuries gave a 4.37% yield, but a 30-year mortgage cost just 6.21%, with 0.5 points.  So government bond rates have fallen 0.57% since then, but mortgage rates have risen 0.31% and 0.2 points.  Not to mention the fact that nobody gets a mortgage anymore without 20% down, a good credit record, and firm proof of employment and assets.  At the beginning of 2006, banks were practically begging you take money, regardless of income or assets.



This difference between one kind of loan and another - in this case between mortgages and government bonds - is known in the trade as a 'spread.'  And by watching these spreads, the pros take the measure of how bad the credit crisis is.  Treasury yields, which have no credit risk (who owns the printing press?), are always the lowest rates, with other rates varying depending on things like credit risk and supply and demand.  So in our example, the 'spread' (ignoring the points) went from 1.84% in January 2006 to 2.47% in January 2007, and is now 2.72%. 

But for the pros, the mortgage rate isn't a good measure, because surveys like Freddie's only come out once a week, and mortgages have all kinds of other stuff - like points and changing fees - wrapped up in them.  When clients asked me what I thought the single best measure of the crisis was, I always said the same thing:  3-month LIBOR vs. 3-month OIS.  In a nutshell, LIBOR - the London Interbank Offered Rate - is the rate that the best banks charge each other for unsecured loans.  'Unsecured' means just what it sounds like - no collateral.  You're just relying on the other guy's creditworthiness and good faith.  This gets published every day by the British Bankers Association.  More history and discussion of this rate will have to wait, but it's easy to find, it comes out every day, and everybody knows what it is.  OIS is a little more complicated, but the 2-cent version of it is that it measures the rate that the Federal Reserve sets for banks to borrow and lend each other's reserves held in accounts at the Fed.  You'll have to take my word for this, but these are just about the safest loans there are.  So measuring the spread between LIBOR and OIS is a measurement of the safest loans against the much more risky unsecured loans.  Wall Streeters like the 3-month rate because it is an extrememly common maturity for this type of loan.

So what has this spread done?



This is a printout from Bloomberg, LP's information service - known as 'the Bloomberg.' If you ever wondered where Mike Bloomberg got all his money, it was from creating and marketing the best tool in the business... but I digress.  Look at this spread.  Before June 2007, it bumped along at about 0.15%, pretty much forever, with occasional fluctuations at quarter-ends and year-ends.  Then look what happened - it went screaming out to 1.00%, then became incredibly volatile, peaked at 1.06%, flew back in when it looked the banks had things under control at the end of 2007, and then ramped back up in March when a certain brokerage went kablooie.   What has some people on Wall Street worried is that despite all the writedowns, all the actions by central banks to make money available, and all the time that's passed, this spread has been sticky and has even begun to move wider again.

If you made it this far, congratulations!  We're near the end.  Want to watch these rates on your own?  It's not easy, but you can get the data if you try.  One way is to go to www.bloomberg.com, enter the ticker symbol ussoc:ind (for 3-month OIS) or us0003m:ind (for 3-month LIBOR) and click on chart.  You won't get the above chart, but you'll get a nice picture anyway. So track it like the pros, and don't let anyone tell you the crisis is over until this spread returns to normal.

3 comments:

  1. Painful. Just painful.

    ReplyDelete
  2. Can you explain the following to a laymen:

    As I understand the OIS is based on the Federal Fund Rate - why is it spread between LIBOR and OIS a better indicator than the spread between LIBOR and Fed Fund Rate?

    And why is the OIS rate currently about 0.5% lower than the Federal Fund Rate?

    thanks - Michael

    ReplyDelete
  3. Michael-
    In answer to part I - The Fed Funds rate itself is an overnight rate, and while there is an overnight LIBOR rate, there's way too much volatility in overnight LIBOR. Furthermore, 3-month LIBOR is a tremendously important benchmark rate - I would guess that upwards of 90% of all floating rate loans are indexed to 3-month LIBOR. So we need a way to compare Fed Funds to 3-month LIBOR. OIS is the way. 3-month OIS is the market's guess about what the AVERAGE Fed Funds rate will be over the next 90 days.

    Here's how it works in practice: OIS stand for overnight index swap and it's just that - a swap. One person takes the fixed side - the OIS rate, which on Friday was 1.45%. The other person takes the floating side. In 90 days, we'll average up all the overnight Fed Funds rates actually seen in the market. If that average is 1.45%, we break even. If it's higher, the floating side wins, to the tune of $25 per 'basis point' (that's .01%) per million dollars in the original OIS contract. So if the two parties did a $1 million swap, and the average rate turns out to be 1.50%, the floating side wins 5 * 25 = $125. Obviously, if the average is lower, the fixed side counterparty wins.

    Now, to the second part: why is OIS lower than the Fed Funds rate. Good question, and there are two parts to the answer. The first part is that because OIS is a guess about the next 3 months, it anticipates changes in the funds rate that might happen over the next 90 days. The market clearly believes that the Fed is going to cut the target rate, and soon. The second part is a little more 'inside baseball.' - OIS doesn't pay off on the average fed funds TARGET rate, the rate that the Fed is aiming for. It pays off on the average Fed Funds EFFECTIVE rate - the weighted average of all Fed Funds loans made during the day. This can fluctuate a long way from the target rate when rate changes are imminent or there is lots of stress in the market. Both factors are coming in to play right now - the last known effective rate is for last Thursday (Friday's will be published Monday morning, and I'm writing this on Sunday, October 5th), and it was 0.67%. By the way, the www.bloomberg.com ticker for the fed funds effective rate is fedl01:ind (that's eff-eee-dee-ell-zero-one). The Fed's target rate is fdtr:ind.

    ReplyDelete