For the first time since announcing its plans to buy ETFs on March 23, The Federal Reserve released detailed ETF holdings and transaction data this past Thursday. While they report aggregate data through May 27, the detailed data discloses ETF holdings and trades through May 18th, (in a separate “Report to Congress” publication). So what does this data reveal?? Drum roll please…
Summary: On May 27, the Fed owned $3.0B in Corporate Bond ETFs. The weekly trend of ownership in ETFs is shown below, for an average daily buy of just under $300 million since the program began on May 12th. By comparison, the full category reported about $14B of inflows in May and trades about $10B daily on exchange. As expected, the purchases were predominantly Investment Grade, with the rest being High Yield.
Trade details of the purchases were released for the first six days of buying, May 12-18th, by which time the Fed had purchased $1.6B in ETFs, The transaction data reveals the daily buying activity below:
If you are going to delve into the details of the transaction data (I wouldn’t recommend it), note that the trades are based Trade Date, while the balance sheet data is based on Settlement (T+2).
Which ETFs were purchased? No major surprises here, as the program stuck to the mandate of buying the largest, broad-based, most liquid ETFs in the IG and HY space, and roughly sticking within existing issuer market share positions. Recall there is a stipulation in the agreement with BlackRock that they are not to buy iShares above their respective market share allocation. Total value of Fed ownership on May 19th is show below:
As expected, ~85% of purchases were Investment Grade, with the remainder in High Yield. If we assume the same IG/HY split continuing through May 27, then the Fed would have owned 1.5% of the IG corporate ETF universe, and 0.8% of High Yield ETFs outstanding.
It may be more interesting to break down Fed ownership between IG & HY:
Within the High Grade category, the purchases were evenly split between short duration, intermediate, and full-curve ETFs, which slightly overweights short-end ETFs vs. the market. As I have mentioned in the past, the Short Duration space makes the most logical sense when it comes to supporting corporations through this crisis.
Addition top-level breakdowns shows the total issuer breakdown for both categories:
The market share figures are based on my estimates, and are subject to interpretation (include target date? active? FRN? ultrashort?) but in any case, the issuer weights look reasonable, and seems to show that BLK made an attempt to lean toward non-iShares products.
How was trade execution? While it would take some serious parsing of tick-level data to determine exact quality of execution, we can look at daily figures for the largest funds purchases to get a sense of trade execution.
In the charts below, I have shown the amount the Fed purchased of the top six funds (80% of total), and overlaid the Fed’s average purchase price that day (Orange), with the High & Low ticks for that day. No surprise, the purchases tend be right around mid price, or sometimes slightly better. I would have been shocked to see anything less.
What about Premiums on days of purchase?
In the charts below, I have shown daily $ purchases (blue, again) overlaid with the closing premium on that day (orange) and the 52 week average premium for that fund (gray). It seems clear that there were some decent bargains to be found in High Yield that week, less so in the well-bid short Investment Grade products. Recall that the Fed’s direction was not to buy funds at premiums that were higher than 1% or one standard deviation above 52wk average.
(ADDED MAY 31): Since the prices that BLK purchased these ETFs at were better than closing prices toward the end of the period, the actual Premiums on the Fed’s trades were slightly better than those reported for funds. See an example of this below, where I show the daily Fund premium (BLUE) vs. the premium based on the Fed’s trade price, with the yellow bars showing the advantage (or disadvantage) of the trades vs. the closing prices:
For a additional reference, see below the premium/discounts on these five funds for the last three months, as well as a YTD price chart for LQD, VCSH, and JNK. (ADDED MAY 31)
What was not purchased? Some ETF corporate types the FED did not wade into so far include Floating Rate, Target-date structures, Active Fixed Income, or Ultrashort ETFs. On the floating-rate side, it might be too sensitive given iShares’ dominance of FLOT in the space, which is 10x bigger than the next sized FRN fund.
There is clearly a lot of interesting data that can be gleaned from the detail trade data, so I expect will be more interesting as we get more data, and also please contact me directly if there is a specific subject or angle of research that intrigues you.
Counterparties: BlackRock used a broad set of approved broker/dealers for the Fed’s ETF transactions, as shown below. Based on the timing of broker approvals [ SMCCF eligible sellers ], I would expect to see larger representation from Citigroup and JP Morgan in upcoming weeks.
So in summary, no major surprises in the ownership data as the Fed stuck to the largest, most liquid ETFs, mostly in Investment Grade, and kept the iShares representation slightly below market. Given the current health of the corporate credit market, one can argue the necessity of these purchases in general, but that is a discussion for another day.
Stay safe everyone.
Elya Schwartzman is the founder and president of ES Investment Consulting LLC (ESIC LLC), an independent advisory firm specializing in ETFs, indexing, fixed income portfolio management, and investment infrastructure / technology.
ADDITIONAL LINKS and RESOURCES:
My original article on the Fed Buying Program, March 27: Fed to Buy IG ETFs, and my updated discussion of Fed HY buying plans: Which HY funds might the Fed buy Don’t want to pat myself on the back too much, but… I nailed it, ha ha.
Federal reserve weekly balance sheet publication: Federal Reserve Balance Sheet
Interesting global perspective on the topic from Deborah Fuhr.
My consulting company website: ESIC LLC.com
DATA SOURCES: Bloomberg, ETFdb.com, Ycharts, Etflogic.com
Disclaimers: Any opinions or views in this post should be assumed to be written by a shelter-in-place lunatic, with access to far too much data, a rudimentary knowledge of excel, and no fact-checking or compliance department whatsoever. Caveat Emptor!
ESIC may now, or in the future, provide advisory services to companies in the ETF ecosystem.