| I have read many articles comparing this year to 1987, or 2008 or other years, withrespect to stock market valuations. Until now, however, I haven't read any comparisons of credit bubbles. The following is from Citibank's Stephen Antczak that does exactly that. Here they are: 1. Valuations With regard to valuations, spreads are still north of the levels seen in '07 in an absolute sense, but to a large extent the extra spread represents compensation for factors that have evolved since '07. These factors include higher dollar prices (typical HG non-fin trades at $109 now vs. $103 in early '07) and lower liquidity, among others. After adjusting for such factors, spreads in many parts of the market are more or less equivalent to '07 levels. In fact, it is not all that difficult to find names that now trade through '07 levels. IBM on-the-run 10-years are trading at 76 bp, or 7 bp through '07 levels (in OAS terms). And KO 10-year benchmarks are now at 65 bp, or 13 bp through '07 levels (Figure 2). But for those of you who worry that these examples may be the exception rather than the rule, we looked at the distribution of the spread change of name matched on-the-run 10-year bonds since Jan '07 (Figure 3). It's almost shocking that the proportion of credits that are tighter now than they were in '07 is only a hair shy of the proportion of those that are wider. What this tells us is that it wasn't really all that hard to find names like IBM and KO – pick out of a hat and 1 in every 2 are trading through '07 spreads. 
2. Liquidity Liquidity, defined as the bid / ask for a given trade size, is far less ample now than it was in the pre-Lehman era. This is in part due to how dramatically dealer balance sheets have shrunk — back then dealer B/S totaled almost 4% of corporate bonds outstanding, relative to only 1% now (Figure 4) But that said, it's very easy to overlook how problematic liquidity was back then. The reason is because everyone seemed to take abundant liquidity as a given back then, and trading strategies were based on this expectation — negative basis trades, CPDOs, SIVs, leveraged loans in TRS form, etc. all fit this profile. But this is certainly not the case anymore. In fact, if investors assume anything about liquidity at this stage, it's that it won't be there when it's needed, in our view. And investment strategies have been adjusted accordingly. To illustrate in real world terms, consider the triple-A CLO market then vs. now. In early '07 triple-A CLOs traded in a 1 to 2 cent market for sizes in the 10 mm to 20 mm range. Now the bid / ask is about 25 cents for the same size. So in this context the advantage obviously goes to '07. But as we know, the ability to actually transfer risk in '07 at these levels was fleeting. 3. Rates Treasury yields are low and the consensus expectation is that they will rise at some point. The problem is that mutual fund holdings of corporate bonds are higher now than they have ever been — 15% of the market (Figure 6) — and if outflows occur in response they could overwhelm dealer balance sheets. We did not have this problem in '07 for three reasons. First, rates were higher than they are now (they averaged 4.6% then vs. 2.7% now), so it wasn't a given that yields would move higher. Second, mutual funds only accounted for 8% of the corporate market, about half as much as now. Third, as noted above, dealer balance sheets were much larger, providing more cushion for risk transfer activity. But that said, how much pressure could we see as a result of a rates-induced selloff? We built a framework to estimate mutual fund flows given various Treasury yields, and we find that a yield of 3.5% by year-end would cause an outflow of about $15 bn. Certainly not good, but potentially manageable, in our view. | 
No comments:
Post a Comment
Keep a civil tongue.