Via Peter Tchir of TF Market Advisors [10],
There has been a lot of talk lately about “seasonal adjustments” and what they actually mean and do for the data.
Reporting today’s forecast in “seasonally adjusted” terms would not be incorrect. The temperature today is almost 30 degrees higher than the average March temperature in NYC, so reporting it as 30 degrees higher than the annual temperature is fine. You instantly know that today is abnormally warm for the time of year. It doesn’t necessarily help you choose what to wear unless you know the monthly and annual averages. It isn’t wrong, but the information isn’t perfect either. If you only had national average temperatures, how would you seasonally adjust today’s NYC temperature? That gets more complicated.
We like “seasonally” adjusted numbers because it “smooths” the data. We don’t get big jumps due to the time of year and we can apply trend lines, etc. to the graphs. There is nothing wrong per se with that, but the adjustments can also mask things in the data. On unemployment, is the BLS adjustment better or worse than what other people model? What variables does it account for? Does it properly account for potential effects of how long a recession has been going. How often does “seasonality” change. In theory, the market could see a -200k number and realize that if normally this time of year it would have been -400k, then it is a good number. It would be a good sign, but it is only +200k for example if the seasonality is consistent. Anyways, enough on that subject. Seasonality isn’t bad, and is useful in many ways, but so is the raw data and trying to figure out if the adjustments make sense or need to be modified a lot due to the particular circumstances at the time (like great warm weather).
The markets are almost all doing well so far this morning. Corporate and financial credit in Europe is significantly tighter, with Main 3 tighter, XOVER 11 tighter, and the Financials CDS index 4 tighter. Partly on the back of some German confidence numbers (which don’t seem to be a leading indicator) and hopes that the EU is becoming less austere. Spain has been told to get to a 5.3% budget deficit, rather than the 4.4% one they reneged on last week. Both sides are maintaining the farce that the 3% target for 2013 will be met.
[11]
Spanish bond yields are higher again, and CDS is unchanged – about the only asset that doesn’t seem to be doing better today. According to Bloomberg, the EFSF will be releasing money to Greece, €5.9 billion in March, €3.3 billion in April, and €5.3 billion in May. That is good, because the ECB has €4.7 billion of GGB bonds maturing in March, and €3.3 billion of GGB bonds maturing in May.
After a flurry of early numbers, we can all wait for the Fed statement. If there was ever a day designed to be waiting for the Fed statement sitting outside at Bryant Park or somewhere, today would be that day. But inside, staring at screens, the market will be closely watching the statement. There is some concern the Fed may acknowledge that inflation is actually a concern for the first time, reducing chances of further QE. As the election gets closer, it will become harder for the Fed to move on QE without compellingly bad data (not just bad in the eyes of the Fed, but to general person tracking the economy). There is still hope that the Fed has been downplaying growth in order to justify QE and will signal its intentions to move before it is too late, and the bank stress tests may be another reason the Fed can justify QE, especially in the mortgage market, where the Fed insisted on scenarios worse than most investors expected.