Per the previous post on the subject of record margin debt, some additional thoughts.
Interest rates are a lot lower now then they were in 2000, and any rigorous analysis of margin debt should take that into account. Any significant lowering of margin interest would be likely to stimulate traders to take on increased debt, so a couple of hundred basis points worth of lower rates might need to be “adjusted out” of the new peak number.
The dollar value of stocks traded is 40% higher today than in 2000. At the stock market peak, about 30 billion S&P 500 shares a month traded hands. Today, we are averaging about 45 billions S&P 500 shares a month. Even after accounting for lower share prices, it would seem that an adjustment for margin debt per dollar value traded is in order.
Given the “black hole” of knowledge about who incurred the debt, and what direction they bet, the “as is” number is useless as a predictor. Assuming that such knowledge could be gained, you would need to have that data going back a sufficient time period to make it statistically useful. Further, you would need to adjust your findings for the impact of margin interest rates and the dollar amount of shares traded.
Without such a rigorous analysis, the fact that a new record was reached is immaterial. Even with a rigorous analysis, it would simply be one more data point of sentiment, which may not add additional information past what we already have.
I do have to give Nick Baker at Bloomers a lot of kudos for answering the question quickly and honestly with an “I don’t know.” Bravo! I’m taking him at face value for keeping short interest in mind the next time he writes a margin piece. Nick, if you’re reading, Born2Code made an excellent point about pattern day traders, as well!
The issue I have with reportage of the margin load is certainly not the expected lack of analytical rigor amongst the mainstream media, no, my issue is the immediate use of it for fearmongering and demagoguery inside the blogosphere, and the lack of any intellectual rigor (other than mortis) that will be applied to the margin data. I can pretty much guarantee that at least eight bloggers and/or pundits will have mentioned the margin record by the end of this weekend, and the over/under on how many of them will mention (1) short selling, (2) day trading, much of which is short selling, (3) margin interest rates, or (4) margin amount per market value of stock traded is two, myself included.
Oh, bother.
7 Comments
I think the big issue is the short-term interest rate and liqudiity differentials between now and then. I’m not so sure if it is necessary to compartmentalize short selling if one is just conducting a comparison between the two eras. Speculation is nothing more than guessing, whether the traders have a bullish bias and tend to borrow to buy (late ’90s) or aggressively short every up move in today’s arbitrage ravaged markets. An inflation- and interest rate-adjusted number would be more useful.
In your last post you mention, “retail versus institutional” as a possible segmentation of the data. I think this is a good idea. The fellow over at stocktiming.com does a good job of tracking institutional moneyflow and he has concluded that institutions have stopped accumulating stock on balance since January. He deduces retail is picking up the tab for everything we have seen since. These are not margin debt numbers, but moneyflow.
In any case, I don’t think it’s a reach to assume that more of that margin debt is institutional than retail. I mean look at the huge bonuses Wall Street just gave out. Surely leverage boosted returns.
Excellent analysis. Your point about the aggregate value of stocks today being 40% above 2000 is one which I suspected yesterday when I first read your analysis of the Reuters article.
Intersting site and great analysis Bill. I shall visit often and reveiw how my readers can benefit from your blog.
Wish you all the best and profitable trading,
OptionPundit
“Speculation is nothing more than guessing,”
i think if you read Livermoore’s and Trader Vic’s books you will change your mind.
Bill, your points are good ones but the answers don’t clear up the picture much.
We should expect margin debt to decline as rates rise, but there is no evidence for that. In fact, over the past 50 years, there is actually a *positive* correlation between changes in the broker call rate and margin debt. So as rates rise, so do margin balances (in general - it is a weal relationship).
Even if rates did affect margin debt balances (which they don’t appear to), the broker call rate is approximately equal to what it was in 2000.
Same goes for market cap. The S&P 500 market cap is around $12.5 trillion now, which again is approximately equal to where it was in 2000. I think it’s a pretty safe guess that most margin loans are taken out against S&P 500 components, so that’s likely the most appropriate number to use.
As for short selling, I don’t believe that’s included in the margin debt figures. When you sell short, you generate “cash”, not a debit, and most brokerage firms segregate short sales into a “type 3″ account. That cash is collateral, however, so it can’t be withdrawn.
Because of that, short sales are also not included in the “free credit” balances reported by the NYSE or Nasdaq. Watching those free credits in comparison to margin debt can give very good signals - it gave a major buy signal in 2002/2003 and again in the summer of 2006 when we saw that there was actually as much or more cash (free credits) than debt in customer accounts.
Now, there is around $39 billion more debt than cash in customer accounts. That’s the worst we’ve seen over the past few years, but it’s a far cry from the spring of 2000, when there was around $130 billion more debt than cash.
Thanks for reading and thanks for commenting!
Note, I never said a word about “market cap.” IMO the proper denominator for normalizing margin debt total should be dollar value of shares traded per month.
Born2code,
In the context in which I made that comment, I was saying that knowing whether leverage is bullish or bearish (long or short selling) probably is not very useful information. I day trade futures full time. Perhaps if you interpret the word “guess” to mean just a shot in the dark, then you are right, successful speculation is more than that. But I meant it in more general way, that no matter how much information you have, the inherent uncertainties in the markets, your opinion on the direction is no better than a guess-albeit an educated one. This is just semantics.
I read all those books you mentioned years ago, but they do not offer much insight into semantics.
One Trackback
[…] Nick Baker hat bei Bloomber vor 3 Tagen den neuen Rekord in Margin Debt ($285.6 Mrd.) zum Anlass genommen und hat (besorgt) darüber geschrieben (NYSE Margin Debt Reaches $285.6 Billion, Topping 2000 Record ). Seine Meldung ist wiederum Anlass für Bill Rempel vom (lesenswerten) Blog NO DooDahs! , einige Gedanken über die Höhe des Margin Debt zu veröffentlichen: More on NYSE Margin Debt. […]