Good Morning. On Thursday afternoon, after several terse phone calls to an ETF company, our custodian, and their trading desk, I was forced to use a phrase I didn't think I'd use again in this business: "Live and learn." I've been managing OPM (other people's money) now for 26 years in the stock market. Over that period of time, a span that included a market crash or two, a handful of wars, terrorist attacks, all kinds of scandals, some bubbles, some booms, and some busts, I thought I'd seen it all. But on Thursday I learned a new important lesson about the workings of Wall Street.
In our business, we primarily utilize stocks and ETFs within our portfolios. The main reason for this is there are no holding period requirements, no trading restrictions, and no mutual fund manager to convince when you want to move a few million bucks around. Some of the other big benefits of ETFs include tax efficiency, a lower cost structure than mutual funds, and transparency. Or so I thought.
On Wednesday morning, our Market Environment Model flipped from yellow to green. This told us to move to long positions in the stock market at the close in our active risk management programs. And since nothing short of a major catastrophe would change the signal, we decided to keep things simple by placing "at the close" orders. This told the computers that we wanted the closing price of the security in question. Since our system is based on closing prices and the market can do all kinds of wild things in the last few minutes of the day, getting the closing price of the ETF we were buying (ProShares UltraPro S&P 500, symbol UPRO) seemed like a good plan. Turns out I was wrong.
Come Thursday morning, I was feeling pretty good about the move. While I have a pretty strong stomach for trading, I really hate being "wrong" immediately on a trade. So, I was glad to see that the market was moving up nicely. But that's when I noticed the problem. My UPRO's were not doing what they were supposed to be doing. Instead of providing 3X the return of the S&P 500, the return was less than 2X. And by the time the market closed on Thursday, they had produced just 1.27X of the S&P. Ouch.
To understand what happened, we have to go back to Wednesday's closing action. At 3:59 pm on Wednesday, the UPRO's traded at $136.65. Less than one minute later, they closed at $137.27 (or +0.45% higher). I didn't think anything of it at the time; I just assumed that the market had popped in the last minute. But unfortunately, that wasn't the case.
No, there were electronic market makers involved that apparently stink at their job. Well, there is actually a much more nefarious view of what happened, but the good folks on Wall Street wouldn't purposefully cheat me on a trade, would they?
One of the little things most people don't understand about ETFs is they have a net asset value that is calculated all day long. Whenever demand or supply gets out of whack, adjustments can be made in order to make sure that the ETF winds up trading very close to that NAV. Well, except in the last seconds of the day, that is.
It turns out that I wasn't the only one moving a few million dollars around on Wednesday and there was a fairly large imbalance of buy orders at the close. If my trades were being handled on a non-electronic exchange, market makers would have spent the milliseconds required to fix the imbalance so that the price of the ETF stayed reasonably close to the NAV. But on ARCA (the old Archipelago Exchange which was bought by what is now the NYSE Group back in 2006) there are no such balancing acts. Nope, in the last minute of the day the exchange just jammed the trades in there.
Another thing to understand is that there are algorithms that search out such imbalances, knowing that there may be an opportunity to exploit. So, with more buyers than sellers of the ETF, ARCA just hit the bids provided by the algos and boom, the price went up nearly 0.5% into the close.
What did ProShares, the originator of the ETF, have to say? In short, it was, "Not our problem." They said their job is to provide the multiple of the index on the NAV of the ETF - not on the actual price. To which I replied, "That's great and all, but I can't buy the NAV price." I then asked what the closing NAV was for the UPRO. "136.26" was the reply. This meant that the good folks at ARCA managed to let traders bid up the UPRO by more than a dollar (or 0.74%) in a matter of seconds. I wanted to know how this was possible and what ProShares is going to do about it. After 30 more minutes of what was almost hostile banter, I hung up figuring that I would soon be looking for another ETF sponsor.
The moral of the story is that buying an ETF with an "at the close" order may sound like a good idea. But in reality, it leaves you open to the games that the computers play on Wall Street. And as everybody knows, these folks would sell their grandmothers for the right price. So, while I am not at all happy about it, "live and learn," is the watchword for the day.
Turning to This Morning ...
The last trading day of a strong month (and May has definitely been strong at +3.56% so far) has a tendency to see some give-back. This appears to be the case in the early going today as European macro data (CPI in France, retail sales in Germany, unemployment in Italy) is driving the action. In short, U.S. futures are following European bourses lower at this time.
Here are the Pre-Market indicators we review each morning before the opening bell ...
Major Foreign Markets:
- Shanghai: -0.72%
- Hong Kong: -0.41%
- Japan: +1.37%
- France: -0.68%
- Germany: -0.67%
- Italy: -0.83%
- Spain: -0.87%
- London: -0.93%
Crude Oil Futures: -$0.71 to $92.90
Gold: +$0.50 to $1412.00
Dollar: higher against the yen, euro, and pound
10-Year Bond Yield: Currently trading at 2.079%
Stock Futures Ahead of Open in U.S. (relative to fair value):
- S&P 500: -5.54
- Dow Jones Industrial Average: -57
- NASDAQ Composite: -11.68
Thought For The Day ... "Do not suffer your good nature to say yes when you ought to say no" -George Washington
Positions in stocks mentioned: none