I preface this post by reminding readers that I do not give investment advice on this blog. I may muse about investment decisions I'm making for clients, or for myself, but nothing I say is ever intended to be advice. Ever. Never. I'm serious.
In July, I wrote the following as part of a post about HCA's LBO:
The bottom line is this. If you own a diversified portfolio of stocks, odds are good one or more of them will become an LBO target, and you will enjoy wonderful profits. If you own a diversified portfolio of corporate bonds, one or more of them will become an LBO target, and you will suffer.
I wrote that knowing that at some point, it would hit the portfolios I manage. Today, rumors are swirling about Alltel, which is among my largest bond holdings.
This brings up two worthwhile questions. First, knowing that a portfolio of corporate bonds is likely to include some LBO targets at some point, why not eliminate your corporate bond holdings? Let's say you have a portfolio of investment-grade bonds which yields 100bps more than a similar portfolio of Treasuries. Now let's say that 10% of your positions wind up being LBO targets, and these bonds all lose 5 points. That's about 50bps in losses on a portfolio that started out yielding 100bps more. And I have to say, even if you were trying to buy bonds that were potential LBO targets, if you own 30-50 names, winding up with 10% actually getting bought out would be tough.
The second is, what to do now if you hold Alltel bonds? The 2012 issue, which is well-traded, has moved about 40bps wider over the last 2 months, all on speculation that Alltel could be a private equity target. So there is a fair-sized LBO possibility already priced in. Contrast that with how other telecom names are trading:
BellSouth (A2/A) +86
Duetsche Telecom (A3/A-) +121
AT&T (A2/A) +82
Vodaphone (A3/A-) +101
Verizon (A3/A) +87
Sprint (Baa3/BBB+) +161
Alltel (before today) (A2/A-) +189
So that implies that if the LBO thing doesn't happen, but another telecom thinks Alltel's wireless network is worth buying, there might be 100bps of tightening here. What's the downside? Since the Alltel thing is all rumor and no fact right now, we can't say what kind of balance sheet the privatized Alltel might have. That being said, B1/B rated Ford Motor Credit 7's of 13 are currently about +310 to the 10-year. That seems like a reasonable limit to your downside.
Another possibility is for private equity to buy Alltel only to dress it up for a sale back to another telecom. If that happens, there might be less leverage involved than in other transactions.
All in all, I'm holding my Alltel bonds. Its risky, but I think the risk/reward is worth it.
Friday, December 29, 2006
I preface this post by reminding readers that I do not give investment advice on this blog. I may muse about investment decisions I'm making for clients, or for myself, but nothing I say is ever intended to be advice. Ever. Never. I'm serious.
1) Admist various worries about interest-only mortgages, few have mentioned the largest interest-only borrower in the U.S... the U.S. Treasury. Treasury bonds pay periodic interest only, with principal paid only at the end of the term.
2) A poll of 68 economists conducted by Bloomberg at the end of 2005 showed a median prediction for the 10-year Treasury yield of 5.00% at the end of 2006. Currently the 10-year is 4.70%. Barring a large change in yield today, this will easily be the most accurate year-end survey Bloomberg since 2000. The survey vs. actual 10-year rate over previous five years?
2004 Survey: 5.04%, Actual (end of '05), 4.39%
2003 Survey: 5.25%, Actual, 4.22%
2002 Survey: 5.00%, Actual, 4.25%
2001 Survey: 5.50%, Actual, 3.82%
2000 Survey: 5.40%, Actual, 5.05%
Median of the same survey for year-end 2007? 4.80%.
3) If current levels hold, the 10-year Treasury will have finished each calendar year within 43bps of the previous year end in 5 of the last 6 years. In the previous 8 years, the 10-year changed in yield by at least 68bps every year and changed by more than 100bps 5 times.
4) Also if current yields hold, it will mark only the second time in 9 years that the 2-year changed in yield by less than 100bps.
Happy new year.
Wednesday, December 27, 2006
The market continues to be confused about the direction of the economy. After we rallied through 4.40% on tens early in December, we've now backed off over 25bps to 4.65% today. We've had two half point sell-offs in the last 5 days: on Friday the 22nd and today. Both were extremely light volume days.
I added some duration today. Not a lot, but I'm just not trusting this sell-off. My bet would be that we rally after January 1. I think the 25bps backup is borne out of technicals, not a real shift in market sentiment.
To this end, I took a look at all the U.S. economic releases of any significance so far in December. We can quibble over what constitutes a significant release, I used any release on which Bloomberg conducted a survey. I count 47 so far this month. 18 have come out stronger (or more inflationary) than in November, and 25 weaker. 4 were unchanged.
The picture isn't getting any clearer, so I'm betting that the recent trading range will hold. That means you're supposed to buy in at the high range and ride it down to the low range yield wise. We'll see if I'm right.
Accrued Interest is back after a great Christmas. Hope every one had a great holiday as well.
The week between Christmas and New Years is always a weird week. Many proprietary buyers (banks, insurance companies, etc.) are focused on getting their books in order for year-end, and are often not doing anything in the market this week. Meanwhile, mutual funds are trying to do a little window dressing before they have to disclose their portfolio holdings. They are also dealing with year-end related withdrawals, be that for IRA distributions, gifting, or taxes. Dealer desks are generally trying to pare inventory as it makes their ROA look better when quarterly and annual reports come out.
What's that mean to investors? If you aren't subject to these artificial constraints, and you are in the market to buy bonds, you might be able to get some deals. Particularly if you can be flexible. Over the last two weeks I've received several calls from traders saying "Tell me where you care on this bond." That translates as "I really really really REALLY REALLY REALLY REALLY want to sell this, you name your price."
On the flip side, if you want to sell something, liquidity blows this week. The same trader who just called you trying to give away his positions isn't too keen on bidding on anything. If you are managing your own portfolio of bonds, here's a tip: figure out what cash you will need at year-end before Thanksgiving. Sell bonds before 12/1. Or else get a margin loan and sell bonds in January. In the municipal market, January 1 is the biggest coupon paying date of the year (along with 7/1), so lots of cash flows into the market that people are looking to reinvest. Combine that with low dealer inventories, and you have a situation where municipals are often very well bid in January.
Thursday, December 21, 2006
Looking at the logs from this website, I see that every day at least a couple people land here looking for how to calculate accrued interest on some interest bearing instrument. Since it’s a slow holiday week, I thought why not actually post the calculation? So the following is a little bond 101.
What is accrued interest?
Bonds generally pay interest every six-months. The interest payment is called a coupon. Whoever is on record as holding the bond on the coupon paying date receives the entire coupon payment regardless of whether they've held the bond for 1 day or for the whole six months.
If that were the end of the story, bonds would gyrate in price based on how close to a coupon date you are, creating all kinds of distortions in the market. To prevent this, bonds trade with accrued interest. Any time a bond trades, the buyer pays the seller a fraction of the upcoming coupon payment, with the fraction being equal to the fraction of the coupon period which has already passed.
How to calculate accrued interest
In order to calculate accrued interest, you must first know what day count fraction (DCF) is to be used. The most common is 30/360, which means that each month is assumed to be 30 days long, and the year is assumed to be 360 days. So if 15 days have passed since the most recent coupon paying day, the accrued interest on a 5% coupon, semi-annual paying bond would be…
15/360 * 5%
That’s 15 days out of a 360 day year, so the fraction of coupon earned is about 4.17% of 5%.
Note that because we've assumed a 30 day month, any time there is a 31st, no interest accrues. Also at the end of February, its possible to have more than 1 day accrued between the 28th or 29th and March 1.
Treasury bonds are done on an actual/actual (sometimes noted as ACT/ACT) basis, which simply means you take the actual number of days that have past and the actual number of days in the year when calculating the fraction.
There are also some conventions where the divisor is 365, which works just like the ACT calculation except leap years are ignored.
Why does the 30/360 convention exist? I've heard different stories, but one reason is that it makes various couponing periods easy to calculate. You can do monthly, quarterly, or semi-annual couponing easily, because 360 divides evenly into 12, 4, or 2. You’d never run into a problem where one period is actually longer than another, resulting in more accrued interest being paid than the coupon! The actual/actual DCF doesn't have that advantage. The 30/360 convention is also easy to calculate by hand, which before the days of Bloomberg was probably helpful.
I’m thinking of running these Bond 101 posts from time to time, so if anyone reading this is interested in the definition of something bond related, please post a comment.
Wednesday, December 20, 2006
Yesterday's post was all about Alpha. A commenter asked me about leverage in bond portfolios, which got me thinking about the other half of the CAPM equation: Beta.
For those who didn't major in finance, the CAPM theory proffers that the return on an asset or portfolio is based on three components: the risk-free return, the general market return (beta), and a residual (alpha). Its fairly easy to understand if you think of a stock portfolio. Let's say you have created a portfolio of 50 stocks all of which are in the S&P 500. Any day in which the S&P 500 is up, you're portfolio is most likely up as well, and vice versa. This sort of general market effect is the beta.
Now let's say that your portfolio seems to be consistently more volatile than the market. I.e., if the market is up 1%, you are usually up 1.2%. We'd say you have a beta of 1.2. If at the end of the year, the S&P is up 20% and you have a beta of 1.2, we'd expect you to be up 24%. Same if the market is down, we'd expect you to be down more. Notice that if the beta concept really holds, then someone with a higher beta only outperforms or underperforms because they've taken more risk. Not because of any skill on the manager's part.
This is why a lot of people in the business focus on alpha, which is the residual return after accounting for market movement and portfolio volatility. Again, if the beta concept holds, then someone with a positive alpha is producing returns over and above the risk taken. That's obviously what everyone wants.
The beta concept is fairly easy for stocks, but more complicated for bonds. We know that over the long-term, stocks tend to rise in price, which compensates investors for their risk. So generally investors want a positive beta to the stock market, to capture this long-run price appreciation tendency.
Ignoring income, bonds do not have a long-term appreciation tendency. For a general bond portfolio to experience price appreciation, interest rates must fall. While we can agree that over the next 50-years, stock prices are highly likely to rise, we can't say that interest rates are highly likely to fall persistently. So it follows that investors would generally want a positive beta to stocks, but would be ambivalent about the beta to rates. If we assume that there is no long-run tendency for rates to rise or fall, then the desired beta is unknown.
In fact, if we assume that you could enjoy the income regardless, you'd probably want a beta of zero. Why be exposed to movements in rates at all?
Getting back to the real world, you can't get the income without taking interest rate risk. Today's yield curve is unusual; in most environments you would be giving up a large chunk of income by eliminating duration risk, whether through hedging or buying short-term bonds. Most of the time, there is a large increase in yield as you move from 0 to 5 years in maturity. Then there is a smaller yield advantage to increasing duration from 5 to 10 years. From 10 to 30 years, the yield pickup is usually relatively small. So you can get most of the income by sticking to 5-10 year maturities, and avoid the volatility of holding longer bonds.
Monday, December 18, 2006
A commenter recently asked me what the best way to add alpha for a bond manager was. Its a very interesting question, and one worth exploring.
There are five basic strategies I see most bond managers employing to try to add alpha. Now, I'm speaking of bond managers who have a mostly taxable investment-grade mandate, and the flexibility to move in and out of various sectors and maturity ranges. They are:
1) Interest rate anticipation. This involves moving duration around to try to time interest rates. I know of very very VERY few managers who make a living doing this successfully. Most guys I know make small interest rate bets hoping to add a little value if they are right but not killing themselves if they are wrong. That's basically what I do as well.
2) Sector rotation. This is where one overweights one sector vs. another. In practice, this is a lot like interest rate anticipation, because certain sectors tend to perform better in different rate environments. Some managers use sector selection as a less risky way to make a bet on rates. For example, if you are bearish on rates, MBS are probably a good sector, while Treasuries are probably the worst sector. You can keep your duration neutral but overweight MBS and outperform if rates rise, but limit your under performance if rates fall.
I do some of this, but its hard to overweight Treasuries for very long, because the negative carry eats you alive. So if I'm bearish on corporates, I'm more likly to just own higher quality names. If I'm bearish on MBS, I'll own more 15-year paper or ARMs vs. 30-year paper.
3) Credit analysis. This is where most bond managers live and die. They overweight credit, try to pick the right names, and win the day on extra carry and spread tightening. People can be more or less aggressive with this, some putting big chunks in high-yield trying to buy into recovery situations. Others just try to find yieldy bonds and any tightening is gravy. Because the overweight corps strategy is so pervasive, any time corporate spreads generally widen out, most fund managers under perform.
I do less with corps than a lot of other managers. I like to pick steady names which have certain long-term, fundamental things going for them. I particularly like names where the company has a motivation to maintain their credit rating. Here is an old post about my credit philosophy.
4) MBS analysis. Something a lot of non-bond people may not realize is that mortgage bonds are the largest sector of the investment grade world. They make up about 35% of the Lehman Aggregate vs. only about 20% for corporates. Interestingly, a large number of investment managers try to index their MBS positions or else just buy in the TBA market. Why I hate TBA is another post for another time. Suffice to say that buying only TBA, in my opinion, is like admitting you can't analyze MBS and are trying to minimize the alpha you subtract in this sector. No one can look me in the eye and tell me that buying TBA is a way to add alpha. No way.
Anyway, I think MBS is the best sector for adding alpha, which I posted about here.
5) Quasi-arbitrage. This strategy has become the favorite of many gigantic mutual funds, in part by necessity. This is where a manager builds a portfolio which has the general credit and interest rate risk of a traditional bond portfolio, but its constructed with a series of derivatives and hedges in addition to traditional bonds. So when PIMCO wants to increase its duration, they may well enter into a futures contract or swaps contract as opposed to actually buying any bonds. And that makes sense, because PIMCO is too damn big to actually go around buying bonds. They are forced to act in the derivatives markets, at least in part, because their monsterous size is such a liability. Take a look at the sector weightings in their flagship Total Return Fund (scroll down a bit). 40% cash? Not really, that's cash or CP held as collateral against derivatives contracts.
Now, I'm not one of these ridiculous chicken little types railing against use of derivatives. Obviously Bill Gross has had plenty of success doing this. My question is why? Why do investors flock to this strategy when there are many more liquid and more transparent managers who are having just as much success? PIMCO and Western Asset Management and Blackrock and the like have become so big, who can say how liquid some of their contracts are? I mean, if you are the market, what happens when you want out of the market? Well, there is no market at that point, is there? The problem is not using derivatives per se, its a matter of being so large that you are the market.
Now, most people use more than one of these strategies. Particularly if they are doing the quasi-arb strategy, because you have to be using the derivatives to be making some kind of bet. But I'd say that every manager has a preferred means of adding alpha, and the good managers try to downplay the areas where they are less likely to add value. For example, I'm focused on MBS analysis first, sector weightings second, and credit analysis third. I spend time on interest rates, but I wait until I have a strong opinion before making a portfolio bet.
Saturday, December 16, 2006
So CPI comes out flat, below the Bloomberg survey of +0.2%. Core also flat. Take that, plus the fact that the market had been off sharply earlier in the week, and you knew a big rally was in the works. So the 10-year up 3/4 and the bond up over a point seemed about right to me.
Then something strange happened. The rally started to lose steam. By mid afternoon, the 10-year was only up 3/8 on the day. By 4PM, the 10-year was flat on the day. WTF? How does such a weak CPI print turn into an unch market?
The explanation I heard from dealer desks was weakness in European and Japanese bond markets. Seems kind of odd to me that Tokyo and Frankfurt would take the U.S. market lower during the late afternoon New York time. That means the sell off would have started around 8PM in Europe and like 3AM in Japan. Activity in those areas would be well known to U.S. Treasury traders before 8:30 New York time.
Anyway, I think it may be that PM's were all so anxious to buy on weakness earlier in the week, that by the time we actually got some bullish data, there was no one left to buy. Its a very bearish signal.
Thursday, December 14, 2006
Hearing that the Federal Reserve may have accidentally released tomorrow's Empire Manufacturing number this afternoon. Its due to be released at 8:30 tomorrow. The figure displayed on the site was 23.13, which is a good bit higher than the Bloomberg survey of 17.7.
Who knows if the 23.13 number is accurate or not. I mean, if they are mistaken about what time to release it, the number could just as well be wrong too.
Anyway, they took the link down already. This might explain why the 10-year is about 3 ticks lower than it was for most of today. Also, we get CPI tomorrow, so even if the Empire Manufacturing figure is 23.13, the CPI result is really more important.
The Fed went ahead and released the Empire Manufacturing number, so the rumor was true.
Rumor #1: There was a rumor circulating around Wall Street yesterday that the out-sized retail sales figure was somehow erroneous and subsequent revisions would bear this out. Never heard what the error might be, and honestly, these rumors float around a lot. The rumor obviously was not widely believed, because the sell off got worse and worse all day.
Rumor #2: Asia is staying out of the U.S. bond market lately. I've heard this from several dealers over the last 2 weeks or so. Asia has been a major reason why interest rates have stayed low over this cycle, and its a popular theory to answer Greenspan's conundrum. So if Asian investors truly are diversifying their assets, this will result in a permanently weaker bid for the U.S. bond market. Its the kind of thing that will be impossible to see on any given day, and even hard to quantify after several years past. But if its true, IF, then there will be pressure on both rates and spreads.
Rumor #3: Money managers came in aggressively to buy on weakness yesterday, according to various desks I talked to. While this might seem doubtful at first, given that the Treasury market kept getting weaker and weaker all day, look below the surface. MBS, swaps, and corps were all tighter yesterday. Traditional money managers and bank portfolios are rarely overweight Treasuries, most make their living on spread product. So when spreads tighten in the face of a very weak Treasury market, it might be a sign that money managers are coming in.
Anecdotally, I talked to one dealer who told me that most of the money managers he talks to are short duration, but still came in to buy yesterday. Why? Its a little of the index conscious attitude I was talking about yesterday.
Let's say your duration is 85% of the index. If a month passes and you do nothing with the portfolio, well then your portfolio is 1 month shorter. Plus you've probably accumulated some cash from normal cash flow. So now instead of 85% you are 81%. What if the market is rallying? You didn't like where rates were before, but the market is going against you and you've allowed the portfolio to drift to an even lower duration than you really wanted. So ironically, many times PM's who are negative on the market are the most aggressive buyers on weakness, because they do not want to allow the drift factor to catch up with them.
Wednesday, December 13, 2006
I've mentioned my fascination with poker in this space before. While I don't play very much myself, I see so many parallels to successful trading and successful poker playing. One of these parallels is knowing which hands to bet, and which hands to lay down.
Let me back up and talk a little about the concept of benchmarking. Professional investment managers live and die by whether they beat an index. So much so that many in the profession decry certain managers as "closet indexers." That is, an investment manager who charges active management fees, but really is just trying to match the index.
Now, I'm not saying that some managers are acting as closet indexers in a cynical attempt to retain clients by staying around the benchmark. Most would agree that while outperformance is OK, underperformance gets you fired. Also, most IM's that I know get paid more on asset growth than on relative performance. So if producing returns that are around the benchmark is enough to bring in new assets, the manager may conclude that its not worth risking underperformance to deviate from the index at all. Obviously such a person is not serving their clients.
But many times the term "closet indexer" is over used. When large investors like pension funds or endowments hire investment managers, they do so in an attempt to fill various buckets. E.g., a large cap, small cap, venture capital, investment-grade bonds, high-yield, etc. These buckets are selected and weighted based on estimations of long-run return and cross correlation patterns. What do they use to estimate return/correlation figures? Indexes.
So if I'm hired to run a Lehman Aggregate strategy for a large endowment, but all I buy are MBS, then regardless of whether I'm outperforming or not, I am not serving my client. Then client asked me to manage a portfolio which would have a beta near 1 to the Lehman Aggregate. Same would be true if I held a very low duration or bought a bunch of high-yield bonds. The portfolio I was hired to create has to fit with various other portfolios, and if I ignore that then performance doesn't matter, I should be fired.
So let's say that I just have no view on rates right now. I've done my research and find arguments for both falling and rising rates to be compelling. But I'm running a bond portfolio, so obviously I'm going to have to make some kind of duration decision. No problem, I pick the index's duration. Am I being a closet indexer? No, because the client has asked me to run a portfolio with a beta to the Agg of around 1. So when I have no view on interest rates, I set my position such that my beta is near 1. It isn't a cop out, its good investment management.
I think of it like a poker hand. If I look at my cards and read all the other players and I just don't think I have a winning hand, I should fold. Too many investment managers want to make a bet just because they think its their job to make bets. Sometimes its just your job to wait for another hand or a different market environment.
Yesterday's market action has me wondering if I've misread the street's expectations. I had figured that a cut was priced in by June, and that any minor change in the statement now would only be a set up for a June cut. In fact, the only real change in the statement was to describe the slowdown in the housing market as "substantial."
But the market gave us a bull steepener, meaning that rates in the short end moved lower by a greater degree than in the long end. As I said last week, a bull flattener means that the market expects Fed cuts sooner rather than later. While I agree with the idea that the Fed may cut 1 to 3 times in 2007, I just do not see that the Fed is setting up a series of cuts in 2007. And having Bernanke's crew publicly state the obvious about the housing market hardly changes my mind.
One idea that I keep coming back to is that the curve will eventually steepen. I believe that will take most of the upside out of the 10-30 year part of the curve. If the market has correctly anticipated the Fed's actions, then the long-end will be priced right first, and the short-end will catch up. I'm building a short-end barbell, with some floating-rate positions and some 5-7 year positions. I'm light in the 20-year area. I have some positions in the 30-year area, as protection against a big bull flattener.
Monday, December 11, 2006
The 2-10 slope flattened 2bps today, and has now flattened the last 4 days except Thursday when the curve was unchanged. The 10-year rallied almost 1/4 point, which should encourage the bond bulls after Friday's rout. I'm hearing that many foreign buyers came in on weakness, which has been a recurring theme for most of the last year or so. I'm sure volume was light with the Fed coming out tomorrow.
The rally gives a bear room to work tomorrow. Had we sold off today, I would have bet on a rally tomorrow just based on technicals. However with the 1/4 point rally on no news, I'm suspicious that today's rally is exaggerated by light volume, and I'm inclined to sell it.
The traders I've talked to are chattering about the Fed scenarios for tomorrow, but when you get down to it, most people expect little change in their language. If that happens, then I doubt the market moves much. If anything, it will disappoint some of the "Cut Now!" crowd and touch off a bear flattener.
But like I said, it sounds like most traders expect no change in statement. I agree with that, so if that's what happens, don't expect much market movement.
Friday, December 08, 2006
On Wednesday, Toll Brothers released earnings, posting a 44% decline in net profit. CEO Robert Toll expressed some optimism, saying that "we may be seeing a floor in some markets..." Toll was panned in some spots for their optimism.
But those who bet on corporate solvency for a living are betting with the home builders. I took a look at YTD change in asset swap spread for the 7 major investment-grade home builders. In each case I used a 10-year issue. The source is Merrill Lynch. A positive number means the bonds have increased in spread, indicating the market thinks there is more risk in the name. A negative number is the opposite, the market thinks the company is a safer bet today than at the beginning of 2006.
Toll Brothers (Baa3/BBB-): +11bps
Pulte Homes (Baa3/BBB): -24bps
Lennar (Baa2/BBB): -3bps
DR Horton (Baa3/BBB-): -9bps
Ryland (Baa3/BBB-): +12bps
MDC (Baa3/BBB-): No Change
Centex (Baa2/BBB): -2bps
So we have two wider, four tighter and one flat. None have moved in a big way YTD, although all are much tighter than where we were in June. Here is a graph of the net change over the course of the year in each.
Here is some perspective. Based on a 6% coupon and 40% recovery, a 10bps widener implies a 1.5% increased chance of default over the course of the next 10-years. So the worst performing home builder in this group, according to the bond market, has only a slightly higher probability of defaulting today than it did at the beginning of the year. Further, most home builders have improved in credit strength in 2006.
What does this say about the larger economy. Maybe not too much. The bears will tell you that housing will drag the economy down because of reduced consumer spending via cash out refis as well as reduced construction-related employment. Credit spreads on home builders aren't saying anything about the cash out refi effect. The bond market is saying that home builders will be able to maintain positive cash flow, but that could be because of a massive reduction in spending.
What the credit spreads are telling you is that the bottom is near. If home builders were going to be unable to sell off their inventory and/or take large losses on inventories, spreads would be wider. If there were to be no market for newly constructed homes for several years, spreads would be wider. The credit market has neither of these views.
Thursday, December 07, 2006
Consolidation on Wall Street has pushed out a lot of smaller brokerage firms. Right here in Baltimore, we used to have multiple prominent regional brokerage firms such as Legg Mason and Alex Brown. Both are gone. Today it is difficult to compete with the cost advantages that firms like Merrill Lynch or Morgan Stanley enjoy.
But there remains one area where regional firms have the advantage over the big shops: municipal bonds. Not necessarily the big billion dollar state deals, which do exist in the muni market, but in the smaller deals which make up the majority of the market.
A firm like Merrill Lynch isn't set up to do a $20 million muni deal. There isn't enough in it to make a difference in the bottom line. Enter the smaller dealers, for whom $10-$50 million deals are just the right size.
Of course, municipal deals and corporate deals are very different. In a corporate deal, the corporate executives have a personal financial incentive to minimize interest cost. In a municipal deal, the municipality's finance people may be morally obligated to seek the best deal, but they have no personal incentive. Which opens the door for unscrupulous bond dealers to use political donations, entertainment, and outright bribes to influence how business is awarded.
There used to be a cadre of smaller dealers in the south referred to as "bond daddies," which were known for operating in a "mutual back scratching" fashion. For me the term conjured images of muni investment bankers and local politicians sitting around smoking giant cigars and sipping mint juleps. At some point the banker offers a generous campaign contribution to the politician and says "Now, don't you think this road needs repaving? I'm sure we could float a bond issue to cover the costs..."
If you talk to anyone at Stephens or Morgan Keegan, they will probably laugh at your quaint story about the old "bond daddy" days. But I hear tell stuff like that still goes on in the muni market. Yes, according to the Wall Street Journal and federal regulators, some bond dealers are still willing to spend money to make money, even if the spending money part involves bribery. And it may be that firms north of the Mason-Dixon line have learned a thing or two about garnering new business the old fashioned way.
Federal regulators are not well equipped to deal with this problem for the same reasons that Merrill isn't well equipped to sell $10 million bond issues. The Feds like going after big targets, but the big boys don't make enough money in the muni business to bother with underhanded dealings. Its the small firm with 10 offices scattered across one state that are more likely to be doing something untoward to win deals. Does the IRS or SEC want to sent investigators out to the 1,000 firms like that to find the 2 or 3 that are guilty? If you are a Federal prosecutor, you don't make your career by going after corporate criminals no one has ever heard of and are guilty of a crime no one cares about.
Here is the bottom line. The muni market is a dank and sometimes seedy place. More retail customers get ripped off when buying munis than any other single product. The fact that each bond is so unique and trades occur in small sizes means that the market is far less efficient than other areas of the bond market. In terms of the whole bribery thing, its a classic example of what happens when you give someone power but no financial incentives. This occurs in all sorts of governmental posts all over the world.
Wednesday, December 06, 2006
The ADP report came out strong today: +158k vs. expectations of +100k. Its spooking the bond market, and I think that's a sign the market was out of breath when the 10-year was at 4.40%.
We're also 2bps flatter between 2's and 10's. Its interesting to think about what various combinations of market direction and curve slope direction mean.
Bull Flattener: If yields on the long end are falling more than those on the short-end, it means the market is pricing in more Fed cuts, but is not pricing in the first cut any sooner.
Bear Flattener: If yields on the short end are rising more than those on the long end, this means the market is delaying the expected first Fed cut, but still expects the same or a similar number of cuts eventually.
Bull Steepener: Here, the short end is rallying more so than the long end, which implies the first cut has been moved forward, but the number of cuts has not changed.
Bear Steepener: If the long end is leading a sell-off, this means that while the Fed may still cut soon, the number of eventual cuts has been reduced.
Tuesday, December 05, 2006
I've recently got a couple questions about non-inversion and range notes. First, some background.
A range note is a bond that pays interest if a specified interest rate remains above or below a certain level and/or remains within a certain range. Most of the ones I've seen lately are structured something like a 6% coupon so long as 3-month LIBOR doesn't go above 8% or some such. Any day where the LIBOR rate is above 8%, no interest accrues. That's just an example, I haven't pulled up a specific issue. I've also seen range notes involving currency exchange rates. Normally the range note is issued by a large bank or other financial institution whose credit is behind the principal payment.
Valuing a range note is fairly straight forward. In the above example where there is a fixed coupon, you have two simple pieces. One is a 6% bullet. The other is a modified 8% LIBOR cap. Instead of a normal cap, this is more of a binomial cap, where you pay 6% every time you are above the cap strike. Either way, normal binomial tree methodology could be used to value the modified cap.
I don't own any range notes of this type. If I wanted to enter into a cap or floor transaction, I could. It almost has to be the case that entering into the derivative transaction yourself is cheaper than doing it via the structured note. I say this because the investment bank doing the range note is, in effect, selling the derivative to you. They wouldn't be doing this unless it was profitable for them.
Now, there may well be cases where you're willing to pay up for the derivative exposure the range note allows. First, it might be a situation where legally an entity can't use derivatives. This is pretty common among public authorities and municipalities. The range note may allow the entity to hedge certain interest rate exposures they would otherwise be unable to. Second, a small investor may not be able to buy the derivative due to size constraints. However, most small investors are not going to be able to properly value the pieces embedded in the structure, so there is considerable risk that small investors get plain ripped off.
A non-inversion note is basically a range note but where the "range" is the slope of the yield curve. I've seen them done based on LIBOR and Treasuries and usually the slope is 2-10 years. So as an example, the non-inversion note would pay 8% any time the slope between 2-10 years is positive, zero when its negative. The ones I've seen usually have a 1 or 2 year fixed period, meaning that the bond cannot be called and the 8% coupon is paid no matter what. After the fixed period, the 8% accrues every day the slope is positive, whereas nothing accrues on days where the slope is negative. So if the bond pays quarterly with a 30/360 accrual schedule, and the slope was negative one day during the quarter, you'd be paid 1.978% for that quarter or a 7.91% annual rate.
I view this more favorably than the straight range notes. First of all, the derivatives needed to reproduce a non-inversion note are not standard, widely traded structures. So to reproduce these bonds you'd need to buy something custom created anyway. Second, I believe there is a natural tendency for the yield curve to be positively sloped, and this is a tendency which will persist over time. That makes betting against inversion an easy call. Has recent foreign purchase activity caused the slope to be flatter than it might have otherwise been? Maybe, but remember that as recently as 2004 the 2-10 slope was over +200bps.
The issues are usually callable anytime after the fixed period is over. In all likelihood, the issuer will call the bonds as soon as they can if the curve has reverted to a normal slope. The bonds will only stay outstanding as long as the slope is negative or close to zero. Take that as a warning for anyone considering buying one of these.
Why are these issued? Most of the non-inversion deals have been done by investment banks or commercial banks. These are entities that abhor an inversion because it compresses their profits. I believe banks view issuing non-inversion notes as an insurance policy against long-term inversion. While historically, periods of inversion are usually less than a year, the banks just cannot afford a longer period of inversion, and are willing to pay for some amount of protection.
Monday, December 04, 2006
While waiting for a meeting to start, I found two great blog posts worth sharing.
First is David Andrew Taylor from Dismally wrote a detailed response to a question yours truly posed to him. David does great work at his blog, which I read daily. Most blogs run by real traders are focused on stocks or commodities which are all fine and good but don't have a lot to do with my area of focus. Currencies, which is what Dismally is all about, have a lot to do with interest rates and inflation. So David's blog and my blog might seem to be worlds apart, but we actually deal in the same topics quite often. David also has more of a Keynesian viewpoint, which serves as a check on my own monetarist views.
Second is a post from Econbrowser on how much foreign flows might be impacting interest rates. The research quoted claims the 10-year is 100bps lower than it would be with zero net foreign purchases. The problem with this kind of research is that zero net purchases isn't too realistic. So the headline of 100bps is a little misleading. For what its worth, I believe that the foreign purchase effect is quite large, I just don't think foreigners are going anywhere anytime soon.
One reader had the following to say about Friday's steepener post...
"When do the Fed's Governors quit talking up the inflation fears and finally talk about the weakening economy ..... when it's too late ? It seems as if they're putting themselves into a non-winning position by yelling "fire" so often "
Here is my thought. I'm assuming by "fire" the reader means warning about inflation. I think the Fed faces a real conundrum. There were two major "mistakes" in the Fed's history. One was in the 1930's, decreasing the money supply at a point where the economy was already weak, thus causing the Great Depression. The second was the 1970's stagflation, where the Fed allowed the money supply to grow to help economic growth while allowing inflation expectations to become entrenched at a high level.
Stagflation is a real risk here, and the Fed knows it. The bursting of the housing market bubble is (and will) weighing on the economy both because of job losses and consumer behavior. But the economy is still awash in liquidity owing to the 1% Fed Funds rate of 2003-2004. Inflation measures continue to creep up, the dollar is weakening, etc.
In order to avoid the mistakes of the 1970's, the Fed will continue to talk very tough about inflation. What they want to avoid more than anything else is a loss of credibility. If the public continues to believe that the Fed is still primarily concerned about inflation, and will do whatever it takes to avoid accelerating inflation. As long as this credibility remains, the Fed can afford to just leave rates where they are and allow a weaker economy to solve the inflation problem.
Once their credibility is lost, they become forced to embark on a 1980's style series of painful rate hikes. No one wants that. So even if they think there are serious risks to the economy, as long as the level of inflation remains elevated, they are going to continue to emphasize the inflation risk.
Friday, December 01, 2006
The 2-10 slope is 6bps steeper (less inverted) on the day, which is the biggest single day move in that slope since September 20, 2005. The slope is currently -10bps after trading as low as -19bps at various points in late November.
A bull steepener (which is where rates fall, but short rates fall faster than long rates) indicates cuts from the Fed will happen sooner than later. To understand this, think about why the curve inverted in the first place. It was because the market believed that a period of economic weakness and Fed cuts would eventually come to fruition. Therefore investing in longer-term rates would allow one to lock in attractive coupons. So while shorter-term bonds are more closely related to the current Fed Funds rate, longer term bonds anticipate cuts. Hence, short-term bonds carry higher rates than long-term bonds.
When the curve steepens, something about the above situation has changed. If the curve is steepening and rates are falling, then it must be that the eventually from above has turned into soon.
Is this right? I've been arguing that the economy is on better footing than the market is indicating. Today's weaker manufacturing numbers don't convince me otherwise. In order for me to be convinced, I'm looking for weaker consumer spending/income figures. Yesterday's initial claims figure isn't encouraging, but I think we need more data points to be worried about a recession.
Then again, I've argued that the downside potential for rates (upside for bond prices) is much greater than the upside for rates. If we do enter into a recession, I believe that the Fed will be forced to cut rates very aggressively, possibly down into the 2% range. This would create a major rally, particularly in the front end. That's why I'm positioned duration neutral and for a steeper yield curve.
Rumors are swirling that Home Depot is in the sights of private equity buyers, and that the bid might not be friendly. While Home Depot stock was only up marginally yesterday, the bonds were getting killed. HD '16's were trading in the +75 range last week, now over +100.
In many ways, a buyout makes some sense. First, you've got a languishing stock. Weighing on Home Depot stock is obviously the housing situation, but the stock had been languishing before this year. There are questions about management, highlighted by the near riot at the most recent annual meeting. If a private equity buyer could wait out the housing downturn and remove the current management group, they might earn a nice profit.
On the other hand, this would be an extremely large transaction, likely over $100 billion. I'm also surprised to hear this deal might be done as a hostile take over. I can't think of any major private equity deals done recently as hostile. I'm sure there was one or two, but not any of the really big deals.
Does that make Home Depot bonds a buy? There are two likely paths for this bond. If there is no buyout, the bonds probably tighten to +80 to +85. They won't make it all the way back to +75 because there will remain this lingering doubt over whether the buyout idea could come back.
If there is an LBO, the bonds could go out another 100-200bps. It all depends on what the new company looks like, which we can't know at this point. Given that Home Depot is rated Aa3/AA, so if they were to move into junk territory, +300 or so, that's +200 from here.
So you have 15-20bps of tightening potential and 200bps of widening potential. In percentage terms, that's upside of around 1.5% and downside of 16%. Some investors would stop right there, unwilling to take risk that is so negatively skewed. But for traders it becomes a question of odds. If you think the odds of the buyout actually coming to fruition is 1%, its a buy. If its 10%, its a push. If its 20%, its a short-sell.
For my money, there are plenty of A-rated corporate names with a little hair on them that have 15bps of tightening potential and offer better odds against widening.
Wednesday, November 29, 2006
Let's talk about some sobering realities in regards to Ford, which was downgraded by Moody's to Caa1 on Monday after borrowing a fresh $18 billion.
1) According to Moody's, since 1983, 21% of issuers rated in the Caa range are in default within one year. 43% of these issues are in default in 3 years, and 79% of Caa issuers are in default in 10-years.
2) Ford has pledged every damn asset they have, including their equity in Ford Motor Credit and Volvo, to secure the new loans. In effect, these assets have been taken away as security for Ford bond holders and given to new creditors.
3) This means that, de facto, Ford bond holders have become subordinate to creditors, and are unlikely to enjoy much recovery.
Now here is the good news. Ford has substantial amounts of free (e.g. non-pledged) cash, between $25 billion and $38 billion depending on who you ask. Fitch says they think cash burn will be around $8 billion next year. So at that rate, they should last another 3 years or so. But then again, if auto sales are weakening, the $8 billion figure may turn out to be conservative.
I talked to a corporate trader today who is heavy into trading Ford bonds. They've now hocked everything they can, he told me. If their current restructuring plan doesn't work, and work quickly, Ford is going to implode.
As Julius Caesar once said, the die is cast.
Tuesday, November 28, 2006
Ben Bernanke spoke today in New York before the National Italian American Foundation on his economic outlook. The speech was mainly void of any relvalations. The Fed is more concerned about inflation right now, but sees inflation as moderating over the next year as the economy slows. We've heard that same basic statement from several Fed speakers as well as the public releases.
He did speak directly on the topic of rising labor costs, which I wrote about a few weeks ago. Here is the quote.
"What implications does the pickup in labor costs have for price inflation? One possible outcome is that increases in labor costs will largely be absorbed by a narrowing of firms' profit margins and not be passed on to consumers in the form of higher prices. The fact that the average markup of prices over unit labor costs is currently high by historical standards suggests some scope for this outcome to occur. If higher labor costs are mostly absorbed by firms and not passed on, then workers will see the gains in their nominal compensation per hour of work translated into greater real compensation per hour; in the process, workers would capture a greater share of the fruits of the high rate of productivity growth seen in recent years. The more worrisome possibility is that tight product markets might allow firms to pass all or part of their higher labor costs through to prices, adding to inflation pressures. The data on costs, margins, and prices in coming months may shed some light on which of these two scenarios is likely to be the better description of events."
So it comes down to your view of pricing power. Until recently, the feeling was that firms had very little pricing power, that globalization was taking it away. But once again, the money supply rules, and now it seems the deluge of money the Fed gave us in 2002-2003 has allowed pricing power to re-emerge. We'll see how it plays out.
Reader HFT asked me about Ecuadorean bonds, so here are my thoughts. First, some background.
Rafael Correa is now claiming victory in Ecuador's presidential election, and has threatened to default on its foreign debt. As ___ from the University of Sussex said in today's WSJ, "Correa wants to press the reset button on Ecuadorean politics..." He is also looking to rewrite the Ecuadorean constitution.
The CDS market for Ecuador is gapping, and is now over 500bps up from 250 at the beginning of the month. According to Bloomberg, that level is wider than Lebanon or Iraq. Now, I'm not an EM trader so I cannot give much insight into whether 500bps represents a buy or what. But its interesting to contrast this situation with that of a regular corporate bond default.
When a corporate bond defaults, bond holders have a legal right to the defaulted company's assets. In a simple situation where bond holders have the only claim in bankruptcy court, bond holders could choose to become the new equity holders of the company, or else could choose to liquidate the assets and take what recovery they could.
The situation with sovereigns is far more complex. There is no legal authority to force a sovereign to pay their debts. When you lend money to a nation, you are really at their government's mercy. The only incentive nations have to pay debt service is continued access to debt markets. As soon as a government hints that it might not repay its debt, the spigot from foreign investors goes dry.
Curious then that the apparent new regime, which has an ambitious social agenda, is so willing to poison the foreign debt well. I read pieces from Citigroup and Merrill Lynch, both saying that Ecuador is in such good fiscal shape that a default would be an odd move. So maybe its all saber rattling. Maybe its Correa's way of communicating to the public just how sweeping his plans are, even if some of the elements of the plan aren't implemented. Who knows?
The question for investors is do you really want to bet on this guy? Just as Europe is starting to realize how bankrupt an idea socialism is, it seems that Latin America is renewing its fascination. Does Correa not realize how fungible capital is today? Does he not see that foreign manufacturers currently operating in Ecuador could move to India or China or Singapore in a heartbeat? Why does he want his country to look more like Cuba and less like Mexico?
For my money, the spread on Ecuadorean CDS could be 1,000bps, I'm still staying away.
A weak durable goods report got the bond market all juiced up this morning, pushing the 10-year below 4.50% briefly. But at 10AM, a better-than-expected Existing Home Sales figure took a bit of the wind out of the market's sail, and the 10-year is now only up 3 ticks.
Now I know this is going to be hard to believe, but existing home sales actually increased in October to 6.24 million up from a revised 6.21 million. We know that there will come a time where housing activity stabilizes. When that might be and at what levels remains to be seen. But looking at existing sales, the last 4 prints have been 6.33, 6.30, 6.21, 6.24.
Seems to me that if existing homes are moving at a consistent pace, then the market has found a clearing price. If that's right, then prices should recover as the new home inventory is worked off.
I don't think four prints from existing home sales is enough to reach the above conclusion, but by the time we actually have enough data to reach that conclusion, the market will have already moved. So I'm putting my money on a soft landing.
Monday, November 27, 2006
The Dow is getting hammered today and I think that's what's keeping the bond market above water. The 10-year was down 10 ticks earlier today on reports that the weekend sales were strong, but with the Dow down 160, bond have rallied and are now +5 ticks on the day. That's a 1/2 point turnaround.
Interesting that the bond market has performed so well the last two trading days given the dollar sell off. If inflation is defined as a decline in purchasing power, then a weaker dollar is synonymous with higher inflation. This is particularly true in the U.S., where such a large percentage of consumer goods are imported. Weaker dollar means the price of imported goods goes up.
A more technical way to look at a dollar sell-off is that storing capital in U.S. dollars has become less attractive. So for example if foreign interest rates are higher than U.S. rates, then investors may choose to invest overseas and enjoy the greater carry. Of course, if this were happening, then we'd still expect the bond market to decline, which isn't happening.
One realistic possibility is that investors are pulling money out of the U.S. stock market and investing it overseas. This seems plausible given the sharp sell-off in U.S. stocks today. But if that were happening, I'd expect European stocks to ourperform the U.S. over the last few days. That's not happening either. The FTSE, the CAC, and the DAX are all underperforming the DJIA since Wednesday. The Nikkei is a little better, but it isn't the yen that's been driving this recent move, its been the euro.
Anyway, I think there will be better entry points in the near future. I'm holding any cash I have until I see 4.65% on the 10-year.
Wednesday, November 22, 2006
A post about being wrong. My wife would love it.
Several times in the last couple months I've made the case for a steepener. So far, so shitty on that bet. The curve has moved persistently more inverted since the end of September, now sitting at -18bps (2's vs. 10's), which is about the most inverted it got in February before making a big bear steepener in the spring.
Even though its painful to be wrong no matter what, at least part of my investment thesis is playing out (so far): that there was a limit to how inverted the curve could become. So having the curve go against me for ~20bps has a pretty nominal impact on relative return. I view my curve bet to be as much about the relative payoff as it is about the odds of a steeper or flatter curve. Consider the following:
1) Periods of curve inversion tend to be short, lasting less than a year. The curve has been inverted or nearly so for almost all of 2006.
2) The 20-yr average for the 2-10 slope is 90bps and the median is 69bps. The maximum (using month-end values) slope is +266, the minimum is -47.
3) The curve tends to steepen after the Fed starts cutting, usually flattening ahead of Fed cuts.
So we know that the inversion is already advanced in age and the upside for slope is greater than the downside. That makes a steepener a pretty good bet just based on risk/reward. If I'm wrong, and we get to the -50bps area in slope, that's only -30bps from here. If I'm right the median figure is +90bps from here.
The third point is the kicker for me. If the Fed starts cutting, there will be an initial period of inversion, followed by a severe steepener. If the Fed doesn't cut, then the curve will at least get back to a slope of +10 to +20 or so, because the inversion that now exists assumes some number of Fed cuts.
Tuesday, November 21, 2006
Blackstone's purchase of Equity Office Properties (EOP) is getting a lot of attention from bond traders for various reasons. First, we had all assumed that REITs were relatively LBO-proof. This is because REITs are generally highly levered to begin with, so piling more leverage on top was thought to be unfeasible in most cases. It is also because most REITs live with fairly restrictive bond covenants. So if a private equity firm wants to buy up a REIT, it isn't easy to do it with debt.
However, the financiers are proving smarter than the lawyers in finding ways around the covenants. Rumors are abound on how Blackstone will handle EOP's covenant debt. They always have the option of doing make-whole calls, but the make whole will be mucho expensive. EOP's 30-year debt has a make-whole call at +30bps to the old 30yr. That would mean those bonds would cost about $1.4 on the dollar to call via a make whole.
They would more likely tender bonds. I.e., Blackstone would publicly offer to buy bonds at some price from existing bond holders. The problem with that is bond holders know that Blackstone is forced to retire debt, therefore there is an incentive for bond holders to not sell. Usually in these cases, the tendering firm (Blackstone) will attempt to have the problematic covenants simply dropped. I'm hearing that EOP's debt is governed by simple majority, which would mean that if Blackstone can convince a majority of bond holders to drop the covenant, there ceases to be a covenant issue and the outstanding bonds can remain outstanding. They don't use charisma to do this, they use cash.
Here is how it might work. Take the 30-year bonds I mentioned earlier which have a make whole call worth $140. Blackstone tells bond holders they will buy bonds at $120 if the holder also agrees to vote to drop the covenant. This creates a sort of prisoner's dilemma. If a bond holder refuses the tender, they might get $140. But if a majority of bond holders accept the tender, the bonds go back to being worth par. Actually, less than par, since the new company would be more highly levered and undoubtedly be rated below investment grade.
Another option for Blackstone would be to "ring fence" certain bonds. This would involve creating a subsidiary with exactly enough assets to meet covenant requirements and assign the most expensive make-whole call bonds to that sub.
So these bond games will be fun to watch if you are an innocent bystander. If you are a cash bond holder, its probably going to be a profitable venture. If you are a CDS holder, the situation is a bit more murky. No default event has occurred, so nominally nothing has changed for the CDS holder, but at the end of all this, what deliverable securities will remain? Let's say that 3/4 of all bonds are tendered and all covenants are dropped, but 1/4 of bonds remain outstanding because they are stuck in some structure. Now those bonds are downgraded to junk, and for CDS holders, those becomes your reference bonds. So whole cash holders win, CDS holders lose.
Another harrowing part of all this is the realization that no one is safe from a buyout. If a highly levered REIT with restrictive bond covenants can command an 8% premium in a buyout, who knows what's next?
Monday, November 20, 2006
If you were focused on the rate gyrations from last week, you might have missed what was actually the most important move in investment grade fixed income. Swap spreads. For our more casual readers, swap spreads represent the difference between the Treasury rate and the rate on the fixed portion of an interest rate swap. So for example, if one wanted to enter into a 10-year interest rate swap today, where you would be paying a fixed rate over the 10-years and receive 3-month LIBOR, the fixed rate would currently be 5.10%. That is currently 50bps over the 10-year. So we would say the 10-year swap spread is 50bps. Swaps are available for pretty much any term you like. Swap spreads tend to be about the same as very strong corporate spreads. On the short end, the corporate spread might be a little tighter and on the long end the swap spread is a little tighter.
This is extremely important for determining spreads of other products, in large part because swaps are a common means of hedging a fixed rate position. So if you are a corporate trader looking to take on credit risk, but want to eliminate interest rate risk, you would use a swap to do it. And if the swap spread rises, then your desire to own the hedged credit position diminishes, causing credit spreads to widen. MBS and Agency debt is also highly swap correlated.
The 10-year swap spread moved 4.5bps tighter in the last week, which might not sound like much, but is actually quite a move. 4.5 bps is about 3/8 of a point on the 10-year. The graph below shows it a little more dramatically.
The swap rate (green) moves down suddenly right at the end. There are a couple broad reasons why swap spreads tend to move. One is the slope of the curve, because a steeper slope allows for more carry trades, which increases the demand for rate hedges. But as the second graph shows, the curve has moved a good bit flatter (more inverted) during this period.
In fact, look back to June, when swap spreads were their widest. The slope has moved consistently flatter/more inverted, but swap spreads keep tightening. So that does not explain it.
Sometimes when the Treasury market sees a big rally, the swaps market lags. From 11/10 to 11/17, the 10-year was 1bps higher in yield. So that isn't it either.
Volatility can move swap spreads, with lower vol implying tighter spreads, but vol isn't moving. The implied vol on 5x5 swaptions was virtually unchanged last week. Vol has climbed from 15.3 in June to 16.8 now. All that time, swap spreads have gotten tighter.
To me, there is only one good explanation. Cash. Cash coming from the sidelines into the market and not enough supply to take it. Think about it: we know that MBS issuance is way down this year due to housing activity and higher rates. We also know that corporate issuance has been lackluster for a couple years. Even Agency issuance has been down due to delevering on the GSE's part. Meanwhile everyone from PM's to CDO managers to foreigners have to put cash someplace.
Tuesday, November 14, 2006
Friday, November 10, 2006
I recently read a Merrill Lynch commentary that mentioned, somewhat off-handedly, that an officially sanctioned inflation target for the Fed is less likely with a Democratic congress. As I've said many times, I'm a big fan of inflation targeting for a variety of reasons. I think most people who are of the Milton Friedman school of monetarism are at least sympathetic to the idea.
Inflation targeting works for various reasons, but one of the primary reasons is wrapped up in rational expectations theory. That is, when people expect a certain level of inflation, they will act accordingly. For example, when negotiating wage contracts, or borrowing rates, the expected inflation level will influence the final figures. These actions will steer inflation toward the expected level.
If you believe in rational expectations, then expanding the money supply can't have any impact on real GDP growth. If the public knows the Fed is providing easy money, they will adjust their inflation expectations. So laborers get raises but firms raise prices all at the same rate, so no one is better off and inflation is higher.
Keynesians reject the notion that economic actors are all that rational. So let's say there is a glut of inventories in the economy at the time the Fed is expanding the money supply. Now consumers have more money but maybe firms don't raise prices right away and instead choose to reduce inventory. Had the influx of money-related demand not occurred, firms may have cut production in order to eliminate inventory, which reduces GDP. But because of the money expansion, the inventory is sold and production levels remain high.
I'm sure most people reading this are thinking both scenarios sound pretty reasonable. I'm more sympathetic to the monetarist perspective. In the second scenario, I agree you'd wind up with inventory reduction, but before too long the easy money would result in pure inflation, and that increased level of inflation would become ingrained in expectations. Now to bring inflation expectations back down, the Fed would need to conduct a painful round of tightening. So the short-term benefit of better GDP through easy money is more than offset by the need for severe tightening down the road.
But my opinion isn't the one that matters. If the people in congress, and the people advising them (like Paul Krugman) think more along the Keynesian line, then they will more often advocate monetary policy based on various economic figures, like capacity utilization and business inventories, not just core PCE. Ben Bernanke would then be in a political fight to get his inflation target, which not only would jeopardize his job should the Democrats take the White House, but could even impact the Fed's independence.
Bernanke can run a de facto inflation targeting Fed anyway. He makes a public announcement that he wants core PCE to be between 1.5% and 2.5%, and promises to do everything in his power to keep it there. Pretty much an inflation target eh? Why risk the political fight?
Thursday, November 09, 2006
In thinking about the Democratic sweep on Tuesday, I continue to struggle with what they are actually going to do with their new-found power. The consensus is they'll try to increase the minimum wage, but I think most serious economists will agree that this will have minimal impact on the economy as a whole. Its also widely reported that the Dems are interested in curbing executive pay.
This gets me thinking about the issue of income disparity. The fairness element of this issue is a debate for another forum. My question is, what are the economic consequences of a widening income disparity?
First, it seems there is no single measure for income disparity. What is the correct way to measure this? Is it the top 10% vs. the bottom 10% in income? Or 1%, or 20%? Is the relevant question how do the richest Americans compare with the poorest, or how the poorest compare with the median, or how the median compare with the richest? Sometimes the "poverty line" is used in calculating income disparity, but this tends to be a moving target.
In my mind, we cannot define how to measure the issue without defining what economic problems might arise. The classic argument is that too much income disparity leads to social unrest. So let's think about that for a moment. If the gap between the wealthiest 10% and the poorest 10% is widening because the rich are getting richer but the poor are about the same, then there is no reason to expect social unrest.
History tells us that social unrest tends to grow out of hopelessness more so that simple disparity. The French and Russian peasants lived with the feudal system for centuries and rose up not because they suddenly resented the rich, but because they were starving. I'm sure they resented their feudal lords all along, but it was only once their situation got worse that actual social unrest occurred.
Now, it sounds like I'm being incredibly callous about poverty in America. This argument is not to say that there aren't people really struggling nor that there is nothing we should do about poverty as a society. Again, fairness is another issue for another blog. What I'm saying is that the issue of income disparity, as it exists in America, is not a macroeconomic growth problem. It is a political issue. A moral issue.
So as January rolls around and the Democrats start making the case for a higher minimum wage and for curbs on executive pay, there will undoubtedly be arguments that there are economic problems with large income disparity. Whatever your position is on these issues from a fairness/moral perspective, income disparity as an economic problem is dubious.
Mixed bag of economic data today. Import prices were low, inventories high, and consumer confidence low. That bolsters the bond bull's position. But jobless claims and the trade deficit were both low, which bolsters the bond bear's position. None of these are numbers I'm all that interested in. We really have to wait until Tuesday of next week to get anything real meaningful with PPI and retail sales. I'm focused on generalized inflation figures and consumer spending, and to a lesser extent, housing inventories.
Wednesday, November 08, 2006
The bond market has been grinding higher the last 3 sessions after last Friday's blood bath. Today I keep seeing headlines saying Treasuries are advancing on the Democratic sweep of the legislative branch. I don't think so. I think you had fast money move the market lower on Friday, but so-called real money accounts pushing it higher the last three days.
I'm basing this on three things. First, I'm hearing from a lot of small dealers that anything in inventory not nailed down is getting lifted. If someone like Morgan Keegan or First Tennessee is seeing their 500 bond Freddie Mac positions lifted left and right, that's banks and smaller money managers, not big hedge funds.
Second, spreads are moving tighter in MBS and corporates. If hedge funds were pushing the market around, they'd be doing it in derivatives, which would leave spreads at best unchanged. Probably wider.
Finally, hedge funds and prop traders (i.e., fast money) like to trade the number. In other words, they are the ones who tend to move the market whenever big economic releases cause a large move, like last Friday. Real money tends to form a generalized economic opinion that doesn't change from day to day, regardless of that day's economic number. So when you I see several days of small moves all in the same direction, its indicative of long-term buyers entering the market.
So what's that telling you? There are two possibilities. Either more money managers are becoming convinced by the bearish economic forecasts, or its simple portfolio squaring. I think the later. The most recent piece of major economic data was positive, so I don't think PM's were convinced of the bearish thesis by anything that's happened in the last two weeks. However, regardless of my position, I'm always looking to square my portfolios. By that I mean, get cash that's accrued to work. If a PM is already bullish on rates, he wouldn't wait around much with cash, jumping in with bids anytime there is a sell-off.
To me that sounds a lot like the last three days.
Every other blog in the world is spending today pontificating about the change in control of the U.S. House (and probably the Senate). Talking about how it will impact the markets beyond the next week or so is speculative at best. I say this because the Democrats ran primarily on a anti-Iraq/Bush platform, not really on any sort of economic policy. At one time, the Democrats were the tax and spend party, promoting transfer programs and heavy regulation. But the last really successful Democratic politician, Bill Clinton, spent his years lowering the deficit, reforming welfare, and getting NAFTA past. So A) I think the focus will remain on Iraq, possibly creating neglect on other issues, and B) We don't know what direction the Democrats are going to head since they never told us.
On this blog, however, we talk about the bond market. Since the budget deficit was an issue that caused at least some typical Republican voters to switch sides in 2006, I thought today's entry could focus on the so-called crowding-out effect.
The crowding-out effect is when government borrowing causes higher interest rates, increasing the cost of private investment activity. With the increased cost, some amount of private investment is forgone, hence deficit spending by the government results in "crowding-out" of private activity. Some argue that deficit spending is therefore economically harmful. They either argue that government spending is inherently inferior to private spending, or that any stimulus from the deficit spending would be offset by the crowding out effect.
We know that, all else being equal, if the government borrows more, interest rates will rise. Its simple supply and demand, the government demands more money (or supplies more bonds) and therefore the price of money (interest rates) must rise. So it would seem as though the crowding out effect is pretty obvious.
However, academics looking for an actual impact on the economy from the crowding out effect struggle to find one. I think this is for a couple reasons. One is, many factors influence interest rates, so trying to isolate just one is very tough. Second, the government has changed their borrowing strategy at various times in history, so at one period they may borrow more long-term and another more short-term, which would have different impacts on interest rates.
I think the biggest reason why the crowding out effect is hard to observe is that the national debt as a percentage of GDP isn't that volatile. Here is a graph of same.
The figure increased a bit during the 1991 recession/Gulf War, decreased during the capital gains tax boom of 1999-2000, and increased again during the Iraqi War. But overall, not that volatile. In addition, government spending does encourage some degree of private investment, e.g., the government awards a defense contract to Boeing, so Boeing expands its production capacity. So combine the relative stability of Federal debt, and the mitigating impact of encouraging some degree of private investment, and it turns out the crowding-out effect doesn't have much effect on interest rates in any given year. I'd say more likely that the crowding out effect has a fair-sized impact, its just fairly constant from year-to-year.
That being said, I think there is merit to the idea that private investment is better for the long-term prospects of the economy than government spending. The reason why savings/investment is important for an economy is to replenish our capital stock (i.e., keep factories in working order), and to expand productive capacity (i.e., build more factories). If we have no investment, then our ability to produce more goods to keep up with population growth is hampered. That would cause per-capita income to decline.
Some government spending programs are for infrastructure, which are important long-term investments. But much of government spending has no positive long-term impact on productive capacity of the economy. For example, when the government buys more tanks, the goal is to defend our borders not to expand productive capacity. Defense is an attempt to maintain the status-quo, and while obviously important, the economy would be better off in the long-run if we didn't have to spend as much. We spend on defense because we have to, not because we want to. Similarly, transfer payments (e.g., social security, medicare), don't expand our productive capacity at all. Mind you, this isn't a political judgment about the merits of transfer payments, rather a statement of fact about the nature of these payments.
So generally its up to private investors to build new factories, invent new products, etc. Its clear that federal deficit spending has some impact on interest rates and therefore crowds out some level of private investment. Plus, you ultimately cannot keep spending more than you're bringing in, the compounding of interest will eat you alive. So on the balance I'm for lowering the deficit. But the deficit problem does get way too much play as an economic issue among both politicians and the media.
Monday, November 06, 2006
I was playing golf with a client about 2 years ago, and as I was lining up a putt, I declared that it was impossible for the housing market to get so expensive that people couldn't afford houses. Home prices are governed by supply and demand like any other market, I reasoned. If the price is too high, demand would fall. Since demand (at the time) was in fact, rising, obviously prices aren't so high that homes were generally unaffordable. Obviously it could become unaffordable for a given consumer, but not for consumers generally.
What I ignored was the possibility of widespread investment in residential real estate. According to the National Association of Realtors, 27.7% of homes bought in 2005 were for investment. Investors have to outbid consumers for properties. It may be that a portion of homes bought by investors wouldn't have sold to a consumer, e.g., a home in such bad condition that its unlivable. In the majority of cases, though, investor demand has shifted the demand curve for homes outward.
So far, housing starts have fallen about 500,000 from their peak in January. Investor demand in 2005 represented about 2.3 million homes, so the lost demand from investors out-weighs the decrease in new supply.
As FRB of Dallas president Richard Fisher said the other day, the Fed really created this housing bubble by holding rates so low for so long. I am increasingly of the belief that investors speculating in the housing market really created the bubble.
Investors will not hold on to houses at a loss. They will either rent them out or sell them at a loss and walk away. The good news for the housing market is that if economic fundamentals continue to be strong, then it should only take 1-2 years to wring out the speculative investors. After that, I'd think the housing market will get back to a more normal pace of appreciation.
Friday, November 03, 2006
On the back of yesterday’s fairly high ULC figure, today’s low unemployment figure has the Treasury market running for cover. The 10-year is down a full point and the 30-year is down 1 and a half.
I can see why this is causing consternation for the bond bulls. With inflation figures still high and unemployment falling, what's the argument for Fed cuts again?
Commenter Richard suggested that my "simple math" from yesterday really isn't that simple. He's right for a variety of reasons. Most obvious is that the ULC calculation might not perfectly reflect total consumer income, such as wealth gains from sale of securities or real estate. Another important point which Richard brought up is the problem of imports. If consumers have 5% more income but use it entirely to buy foreign goods, the impact on U.S. inflation may be minimal.
I spoke with John Burger of Loyola College in Maryland's economics department last night about ULC. He supposed the negative serial correlation issue I mentioned yesterday is probably due to poor measurement on the part of the BLS. Be that flawed methodology or that the data is just hard to get a handle on, either way I'm suspicious of the downward trend in ULC.
Thursday, November 02, 2006
FRB of Dallas president Richard Fisher spoke in New York today, speaking mostly about data quality and the need to look at activity beyond U.S. borders when making decisions about domestic monetary policy.
This speech sounds a little less hawkish than other's he's given recently. His (and my) favorite inflation gauge -- trimmed mean PCE -- printed a little lower last month, falling to 1.7% from 2.7% the previous month. The YoY number is still 2.6%, which everyone would agree is too high. Anyway, Fisher characterized the inflation picture thusly:
"It is possible that the trend in overall consumer inflation has peaked and is finally heading lower. Next, the not-so-good news: The overall inflation trend remains at a level above my comfort zone. I am encouraged by the change in direction of trend inflation, and I hope that in the future my CEO and CFO contacts will be telling me that the competitive forces of globalization have kept their pricing power limited or nonexistent."
I have to read that as Fisher would start to feel more amiable to cuts if trimmed PCE continued to print below 2%, particularly if the YoY number would fall to that level.
Unit Labor Costs, one of the stats I view as critical in the analysis of inflation, came in higher than expected this morning: +3.8% vs. survey of 3.4%. Productivity was also below expectations: 0.0% vs. 1.0% surveyed. ULC for 2Q was also revised higher from 4.9% to 5.4%.
The bond market is mildly lower: down about 1/8. I'm surprised its not down more. My suspicion is that the bond bulls are focusing on the trend for ULC more so than the number itself. Witness the graph! The green line is the stat and the yellow is a four quarter moving average.
So while the 3Q ULC rise is well above the 15-year average of 1.74%, it has also decelerated two quarters in a row. Visually, it doesn't look like this is a figure that goes through short-term trends. In other words, it seems to accelerate one quarter then decelerate the other. In fact, only 26% of periods did the change in ULC show the same sign two quarters in a row.
The moving average is more telling. Note that the ULC is calculated on a quarterly basis and then annualized, so a four quarter moving average gives us a good full-year picture. The MA shows more apparent trends. During the deflation scare of 2001-2002, we did indeed see weak ULC figures. Today, we are obviously on an upswing. In fact, the MA is currently 5.30%, matched in this time series only by 5.35% in 3Q 2000.
If the dollars paid to laborers are rising at a 5.3% clip, the marginal propensity to consume is about 100% (i.e. consumers spend every additional dollar they earn), and output of consumer goods is growing below 5.3%, then consumer prices must increase. Even a monetarist such as myself can agree to this more Keynesian approach, because its simple math.
If you hold goods available for sale constant, but increase the dollars consumers have to spend, then the price of those goods must rise. It follows that if you increase the dollars consumers have to spend at a faster pace than output increases, then prices must rise.
Monetarists believe that inflation is always and everywhere a monetary phenomenon. But in real life, its tough to tell what the relevant money supply is. For example, if the Fed puts more money into circulation, but some foreign central bank decides to put an equal amount of dollars into a vault to maintain a currency peg, money available to the economy hasn't really changed at all. So we're forced to look at indicators of increasing money supply, such as rising labor costs.
In real life, my simple Keynesian math example gets a little complicated, because you have to consider availability of imports, which are fluid. But in my mind, as long as ULC is growing faster than 3% or so, core PCE will stay above the Fed's comfort zone. I can’t see the Fed cutting in this environment.
Wednesday, November 01, 2006
The Chicago Mercantile Exchange plans to list futures on credit defaults. It sounds to me as though the process will be similar to a credit-default swap, but will be exchange traded, which should improve liquidity.
A credit-default swap is where one party agrees to pay a fixed payment and in exchange, the other party agrees to buy a reference obligation at par in the event of a default. So let's say you owned some Clear Channel Communications bonds and you wanted to eliminate the credit risk, but for whatever reason, you didn't want to sell the bonds. You sell a credit default swap, and if Clear Channel defaults, you simply sell your bonds to the other party at par. You pay for the protection in the form of a fee, which in practice is similar to the LIBOR spread of a floating-rate bond of the same issuer.
CDS have come to dominate corporate bond trading, because its a cleaner way of achieving credit exposure. Rather than try to find a specific bond, which may or may not be available, you simply short the CDS, which pays you the spread you wanted anyway. CDS can also be more liquid than specific bond issues, which makes it a good place for hedge funds to speculate on movements in the credit market.
If the CME goes through with their program, it would make derivatives trading all the more attractive over cash corporates for large institutional investors. Will anyone actually trade bonds anymore??
Monday, October 30, 2006
It sure feels like there are two camps about where the economy is going, and incoming data is only causing parties in both camps to harden their positions.
Today we have personal income and personal spending. Personal income growth for September came in a fairly strong +0.5%. Incomes are up a healthy 6.8% YoY (3.9% in real terms), which is well above the 20-year average of 5.5% (or 2.8% in real terms). Those thinking the economy is still strong and inflation is a concern can point to strong income growth.
Consumer spending is a different story. It grew a meager 0.1% for September. Consumer spending has risen 5.5% over the last 12-months, below the 20-year average of 6%. Real consumer spending is up 3.4% YoY, which is right on the long-term average.
I look at this data and see at worst a modest slowdown. Consumers spend about a year paying down some of the debt they incurred during the housing boom, then its back to spending as usual. The savings rate right now is -0.2%. If we assume consumers "want" to have a savings rate around 1%, income is growing at 7% (nominal) per year, spending can grow at a 5.8% (nominal) rate over the next year, then resume growing at 7% thereafter. That'd be around 3% spending growth in real terms, only about 0.4% weaker than the long-term average. If we take the rule of thumb that 70% of the economy is consumer driven, that's about a 0.3% ding against real GDP.
I don't see the Fed cutting if this is how it plays out as I just described. But that description is a little overly simplistic. We know there are some housing-related challenges looming. The Calculated Risk blog argues that the impact from weak housing is yet to be felt, both because construction job losses will accelerate and because mortgage equity withdrawal (MEW) with continue to decline.
How far MEW falls depends entirely on how severe the housing correction turns out to be. To me, if income continues to climb, the glut of home inventories will work it self out without needing deep declines in home prices. So MEW settles in around +$200 billion, which is where it was in the late 1990's. That's a decline of around $300 billion from where it was in the 2nd quarter, or 2.2% of GDP. I'm using Federal Reserve figures here: estimates of MEW vary greatly. Anyway, the Fed thinks about half of MEW feeds into GDP, so a decline in MEW probably weighs on GDP to the tune of 1.1%.
Would the Fed cut in response to a 1.1% decline in GDP? Probably, but if they estimated that the impact was a one-off, its not going to be an aggressive campaign. In other words, if the impact of MEW is -1.1% in 2007 but its clear the housing market has stabilized, income and spending are growing, I can't see the Fed cutting more than 50bps or so. To do more would risk inflation at a time when the economy is on an upswing.