E-piphany

As we move into 2023, one of my New Year’s resolutions is to write more frequently on the blog and post podcasts more frequently. I have a list of topics that is certainly long enough. When I was writing commentary for Bankers Trust, and for Barclays, and for Natixis, I wrote every day and somehow I never ran out of words…

Sometimes, as with today’s article, I am going to refer to pictures and observations that I have previously made on the private/subscription Twitter channel. You can subscribe to the Private Twitter feed at https://inflationguy.blog/shop/ . Not only that, but as of January 2023 I have marked the price down from $99 to only $69, which is a 30% nominal decline in the subscription price – and a 35% or so real decline. (Those of you who subscribed at the $99 price unfortunately will have to cancel and re-subscribe to get the lower price because there’s no way for me to edit a recurring subscription’s price, which annoys me as much as it annoys you but I suppose it’s to keep unscrupulous sellers from raising the price without your permission).

Today I want to present some oldie-but-goodie charts that I developed years ago to look at the seasonality of inflation breakevens. In updating the charts, what was amazing is that…the seasonality hasn’t changed much. Fairly consistently, breakevens rise in the early part of the year, and then decline from May to October. It’s not a guarantee,[1] but it is a pretty consistent tendency. The chart below shows, in black, the percentage of the time (1999-2021, so 22 years of history) in which 10-year breakevens increased in the 60 days following that date. So, on January 3^rd, the number was about 70% which means that in 70% of those years, breakevens were higher 60 days after January 3^rd than on January 3^rd. The average increase (including years in which it decreased) is in red, and shows about 10bps on average. That doesn’t sound like much, but it’s an average of over 22 years. Buying breakevens early in the year is typically a good idea.

The next chart steps back and shows the average for the full year, properly de-trending the data so that any drift over time falls out (since breakevens have gone basically nowhere for a quarter-century, this doesn’t do much but it’s the right way). So, breakevens start the year below the level that will subsequently be the average, and by May they’re well above that level. Ergo, it has historically been good to be long into the first part of May. And then I guess you sell in May and go away, to coin a phrase.

None of this is guaranteed, as I said, but seasonal patterns which are consistent are valuable tools. The way I look at seasonals is that I want to see a move of some decent economic value, but mainly I want to see the consistency. And personally I won’t do a trade just to take advantage of the seasonal trend, but if I want to sell and the market shows a strong tendency to rally then I might consider “flat” the same as selling in that environment. Conversely, a market which has a strong tendency to rally when I want to buy is likely to make me be more aggressive getting in rather than trying to steal a tick on the bid/offer by hanging out on the bid. If you’re bearish on breakevens, then I don’t think you should be a buyer just because it’s a good time of the year to buy. But between the low level of breakevens, and the seasonal trend itself…I would be cautious about being aggressively short.

[1] …and some of it is an artifact: in the early part of the year, a breakeven buyer often has negative carry from bad inflation prints in November and December; as that carry passes, breakevens rise. But this only explains part of the early-season seasonality, not the whole thing.