The T Report: Portugal, Diffusion Indices, and QE

Posted by on Oct 3, 2012 in Uncategorized | No Comments


Portugal managed to exchange some bonds today. The government bought back some bonds due to mature in 2013 and exchanged them for bonds due to mature in 2015. This is interesting. I think it has helped risk in Europe as it is a sign that Portugal is not about to default any time soon and doesn’t seem headed down the PSI path.

This appears to have been a purely private transaction. I remain bearish, but this is definitely a positive development. It shows some willingness to work within the normal constraints of the market much more than the Greek PSI methodology.

The key here is to see if the ECB is willing to play along. In the best case, this is the first step towards the ECB being able to extend maturity on its various SMP holdings. More of a dream than a real thought, but I am watching that.

This may explain why European CDS is doing so well, particularly financials, compared to other asset classes.

In spite of almost doubling from the lows, Greek PSI bonds still seem to offer some value. I don’t see any serious restructuring of these bonds without also getting some public sector debt forgiveness or meaningful maturity extensions and coupon reductions. At these prices, that still gives a lot of upside to these bonds, certainly relative to anything else out there.

Diffusion Indices and Libor

We are getting some benefit from certain economic data, which reminds me of my concerns about diffusion indices to begin with, and some special additional thoughts.

The biggest problem I have with diffusion indices is that their design ensures some volatility. Basically it is a comparison to the prior month. Ignoring all the seasonal adjustments, there is a propensity for these indices to revert. It is hard for every month to be better or worse than the last month for extended periods. The problem with diffusion indices is that a series of 42, 42, 44, 52 doesn’t tell us much. The 52 means this month is better than last month, clearly good, but at some point inevitable (the dead cat bounce). It doesn’t tell us much about the overall level of activity. If a stock market went from 8,000 to 7,500 to 7,000 to 6,500 to 6,700 we wouldn’t be particularly excited. Clearly 6,700 is better than 6,500, but it is clear how much worse the market is than when it started. Diffusion indices obscure that.

Then there is the fact that there tends to be little (no weighting) based on size. 1 big company laying off 1,000 workers counts the same as 1 small company hiring 1 worker. It doesn’t invalidate the results, but is worth thinking about, especially when economies are going through structural changes.

Finally, there is the LIBOR effect. Most of the diffusion indices are surveys. They aren’t based on statistics. No one submits last month’s actual data versus this month’s data. They are subjective, and anything that is subjective can have a lot of bias, particularly if the people submitting the data have an economic interest in the outcome. Anyone filling in the questionnaire that has stock options may well be biased to make things seem better than they are. Possibly far-fetched, but we have seen time and again that people take steps to maximize their own economic interest and filling in a survey ranks pretty low under the moral obligation.

So the bump in diffusion indices isn’t particularly concerning to my bear case, though obviously worse numbers would be better. We need to see if a real trend develops and also some consistency since the data has been mixed and given the high degree of uncertainly around the accuracy of the data and the seasonal adjustments, it is hard to tell if all we have is noise here or a real change.

Where Will the QE Money Go?

This remains one of my biggest concerns and questions (right behind whether OMT will launch or fail).

I see the Fed MBS purchases as being “money creation”. While technically that may or may not be the case, from a practical standpoint, it looks to me like we will see $45 billion of new money per month.

I don’t count the “roll” money as new money. One asset is being redeemed and another is being created and purchased. The argument that the MBS purchases will chase money to riskier assets doesn’t seem to apply to the “roll” portfolio in the same way as it does to the new purchases.

So thinking about this more from an accounting perspective, we have one side which is money and the other side which is assets. A huge simplification, but I think a helpful one.

So, in theory all the money should equal all the assets. As money comes in and that grows than the value of the assets should increase. That is my simple view and I think goes along with how most people now view the QE programs.

So in a year we would have over $500 billion of new money, but right now we have 0, and next month it will be only $45 billion. If we are going to work under the actual money theory, then we can’t give it too much benefit until it arrives.

Okay, so this $45 billion comes in, and assets have to increase in value by $45 billion. Okay, but it doesn’t mean that all assets have to go up. What if equities have already risen in anticipation of the money? If there is no more money for equities, and the first wave of Fed money will be used to rebalance portfolios and benefit underweight sectors? Maybe it is EM stocks that will get the new money? The pool of money and pool of assets is global. Maybe it will be commodities?

So the first question is, even if assets have to increase in total value because of QE, does that increase have to benefit stocks? In the short term, probably not, especially if stocks are already unbalanced. Again, back to my analogy of a dam from the other day, at some point this will happen, but in the short run, there are many assets that can attract the new money to balance the ledger and it doesn’t have to be U.S. stocks.

Then there is supply. The U.S. is running a huge deficit. Europe is running a deficit. To fund those deficits new debt is created each and every month. That increases the asset side. If the U.S. is adding about $80 billion a month to its debt, then there is supply on the asset side. Is QE there more to ensure a balance between money and assets, because otherwise if we saw the asset side growing (I’m including treasuries here) without growth in the money side asset prices would have to drop.

This is such a complete simplification that I’m almost embarrassed to write it. Obviously there are many flaws with the thought process, yet, at the same time, figuring out what can happen during QE is important and much of the prior work seems to assume correlation is causation, so it is worth thinking about in alternative ways.


Twitter: @TFMkts