As I was watching equities make yet new highs last week despite lower stimulus prospects, increased Presidential turmoil, and weaker data (albeit still constructive), I thought about the ETF phenomenon. I had mentioned ETFs casually in previous CAs as a source of equity support. Consider all the 401(k)s, retirement funds, and personal investments that are tied up in passively managed funds. A lot of that money is just automatically invested at regular intervals. Workers generally only make allocation changes a “few” times a year. Since the flow of funds into the ETFs are very regular, there are probably funds that sell longer dated puts as a way to buy future equities and generate additional income. This could explain some of the vol (vix) collapse.
As I saw equities continuing to grind higher, I saw fixed income continuing to grind higher. I had previously mentioned there being fixed income ETFs as well. The similar forces blindly buying Exxon at a 35 P/E are buying tens at 2.25. However, there are some key differences between equities and fixed income:
- Fixed income eventually converges to some function of Fed probabilities. Equities can literally be anywhere and there is nothing that equities need to converge to. Fixed income on the other hand, is all about where the future short rates will be over time. That is all about the Fed. The Fed plans on hiking in the near future, while staying accommodative. That is going to put upward pressure on yields. So there is much more room for equities to get out-of-line than fixed income.
- We know of no one who is sitting on $4.5 trillion of equities, and have announced that they will reduce this amount by up to 50% in coming years. Fixed income on the other hand, should start seeing the effects of decreased demand later this year (barring economic weakness). While the Fed is not directly selling their holdings, the relative outcome from their lack of buying could result in higher yields and more volatility.
- On the off-chance we have a higher wage (or inflation) issue, fixed income starts looking downright ugly. Tens at 2.25% don’t look so hot when wages are growing at 3%. Depending on the narrative, higher wages could be good or bad for equities. I realize when tens in Japan are 0.04% and in Germany are 0.33%, 2.25% in the US looks like free money. But headline CPI in the US is at 2.2%. Are people really saying, “yum! I can lock in my money for 10 years around the current headline inflation rate!”?
As we get deeper into the Baby Boomer retirement, we could see a rotation of assets from equities into fixed income. However, the three factors listed above may keep people from doing so more aggressively. Tens at 2.25% make sense if you see economic weakness ahead. I’m just not sure I see it when business growth is strong. And growth could be stronger if Trump finally gets his act together. I’m not holding my breath.
At this point, because of ETF buying, foreign buying, retirement buying, low inflation buying, and just buying buying (i.e month-end buying, Friday buying, summer buying, etc), I’m somewhat dubious that we can get a large selloff (in either fixed income or equities) without a major catalyst. I think when we do, it can be really huge. This is because there are probably a lot of momentum algos that have jumped in on the up-move. When things turn around, they will be one of the first to jump out and go the other way. Between equities and fixed income, the rally in the latter makes less sense to me from a valuation perspective.
Of course, this all depends on the data, and we have some potential catalysts next week. We get PCE and the Employment Report along with some other key pieces of data. The markets are already expecting yoy core PCE inflation to drop to 1.5%. So that’s going to keep fixed income supported. But that should be somewhat priced in already, and after that we get the Employment Report. I have no idea how the data will turn out, but on the margin, I probably think we could get some squaring up on large longs in ten year Treasuries before the Employment Report. And then it’s all up to the data.