Government Reports from Big-Money Investors: Three Things You Need to Know (but don’t)

Four times a year big-time money managers are required to file form 13F with the SEC.  This always sparks news stories naming the most important investors, people like George Soros and Warren Buffett, and drawing conclusions about what they are doing.  The implication is that you might benefit from looking over their shoulders.

You won’t!  The information is worse than worthless — it is misleading.  Here is why, the three things you should know:

  1. The reports are old news.  The law provides 45 days to file after the end of the quarter and there is no incentive to be early.  The process for filing has been streamlined for the modern age, but the requirements have not.  This delay is an eternity in the modern investing world.  The consumer of the information has no idea whether the positions are still valid.  Has recent news been important?  Could the firm reporting be selling into the strength generated by the report?  They have no obligation beyond the legal reporting.  They are free (and should be) to trade in the best interest of their investors as they get new information and opportunities.
  2. The reports cover only long positions.  There is no requirement to report “shorts.”  The implications of this gap invalidate the reports.  Bill Ackman, for example, is widely known for his short position in Herbalife (HLF) and his very public attacks on the company and its business model.

If you relied upon the government to inform you about Mr. Ackman’s short positions, the 13F would tell you absolutely nothing!

Ackman Longs

Here is another example.  George Soros reported long positions in Barrick Gold and a call (a bullish position) in a gold ETF.  What do we know from this?  Nothing at all.  He might actually have a neutral gold position like a pairs trade, long Barrick and short another gold stock that he believes to be weaker.  We don’t know, because shorts are not reported.

His long call position in the gold ETF might be paired with a short call.  Whether the overall position is long or short depends on which strike and expiration date was bought and sold.  Once again, we know nothing about the overall position.  I do not know from the filing whether Mr. Soros is really bullish on gold, and neither do you.

3. The report on options positions is  — can’t think of a kind word — clueless.  Since the government will not approve a method of analyzing an options position, they require something that is really stupid.  The filer reports long options only.  This means that there are no spreads, including both long and short options, even though that is the most common method of trading for big-time investors.  Worse yet, the long options are not described in terms of their actual value.  The value of each option is assigned the nominal value of the underlying stock!!  Professional traders, and the Nobel-Prize winning options modelers, know that an option has a value based upon a variety of factors, including the stock price, the strike price, the time to expiration, interest rates, expected volatility, and expected dividends.  The option has a theoretical value based upon these factors and a “delta” (the expected change in option value for each dollar move in the underlying).

I understand the government problem in assigning a “theoretical value” and assuming some level of expected volatility.  That does not excuse these blunders:

  • Ignoring short positions in the spread
  • Assigning the underlying stock value to options, even those that are far out-of-the money

Here is a great example from Mike Saltzman, my top researcher, associate portfolio manager, and a veteran options trader.

The reported SPY put position (a bearish bet) is 2.1 million.  The government filings multiply that times the value of the underlying spy, reaching a total value of $430 million or so.  Since the total number of puts is greater than last quarter, this is seized upon in the popular media.

In reality, we have no idea of the strike or the time to expiration for these puts.  The implication is that we have no idea of the value or the deltas for each put.  If far out of the money, they might be cheap protection.  More likely they represent a spread, the sort that a professional trader might buy as cheap downside protection.

Suppose, for example, that you did this spread.

Buy 1.05 million Jun’16 180P for .21

Sell 2.1 million Jun’16 175P for  .12

Buy 1.05 million Jun’16 170P for .09

This is a put butterfly, an extremely common  limited risk position. We own the same number of long and short option positions, so risk is limited. This particular butterfly would cost about $.06 in option, and $6 in commissions. It has a lot of potential. If the SPY goes down to 175 it would make  $9.94 per spread or almost  This spread has a very small short delta component.  It will only make money if the SPY were to fall  more than 10% in just a few weeks. It costs eighteen cents (or about 180K) and might make almost $10 million.

This is not really a short bet on the market.  It is downside protection purchased on a risk/reward basis.

Meanwhile, on the 13F this would show up as being long 2.1 million puts, with a value of (think short value) of $430 million — completely unrelated to the actual position value or properties.  There are many other examples of spreads that would fit the 13F filing,  including some that actually are bullish plays on SPY.

From the filing itself we cannot even conclude that Mr. Soros has a short position in SPY.  It is extremely unlikely that he simply bought 2 million puts without any offsetting short puts.  Professional traders usually work with spreads.


The 13F report is an unhelpful and costly exercise.  Those who take it seriously may well do the wrong thing.

I wrote about this two years ago but the media coverage has not improved.  The best investment advice is to ignore these stories.


The Misleading Story about George Soros

A seriously flawed story about George Soros got a lot of media coverage today.  In the midst of options expiration and breaking news from Ukraine, this story might not seem very important.

It was big enough to earn a play from all of the major news sources and a feature on CNBC.

No one got the story right!

I am not going to list all of the errors.  If anyone had exposed the problem they would be in line for my Silver Bullet award!  Instead, everyone seems to have lamely tagged along with some anonymous sources.  I hope no investors traded on this “information.”

The Basic Claim

The news story was that George Soros, one of the most successful investors in history, had made a big bet on the market moving lower.  That the bet was much larger than in past filings (up over 600%!!).  That what he had risked was the equivalent of 17% of his fund.  Etc.

The Reality

None of these claims can be supported from the actual filing.  Here are the problems.  Those that should be obvious to anyone:

  • The information need not be filed for 45 days.  In the era of high frequency trading, this is ancient history.
  • There is no way to determine whether the positions are hedges against other holdings, nor to evaluate the relative size.

These items might require an options professional, but there are plenty of them around who would be happy to comment:

  •  The reports are based upon “notional value.”  This is a calculation of the number of option contracts multiplied by 100 and by the value of the underlying security.
  • The notional value does not represent how much was spent on the option, so using it to show the percentage of the portfolio invested is completely inaccurate.
  • The report does not show which strike price was used.  If buying puts, you could buy some that were 10% or more out of the money (tail risk?) and the notional value would be the same as if you bought puts that were deep in the money.
  • Put another way, the report does not show the “delta” of the options — how much the options change in price given a change in the underlying security.
  • The 13F filing does not require disclosure of short positions, including short puts.  Short puts are actually a long position, so you can see the confusion.  The reported options could be part of a complex spread — and probably are.  The spread might not even represent short deltas.  We simply do not know.

To summarize:  Soros might have a boatload of cheap tail risk puts.  He might have a put spread, including one that might actually lean long.  He could have some kind of volatility play, since he reports both long calls and puts.  He might have taken the entire position off weeks ago.

Do we really care about this?  You probably should not be influenced by George Soros’s position from 45 days ago, but it gets played as a big story.

What has gone wrong?

Our news sources completely failed on this story.  What went wrong?

Budget cuts?  Summer vacations?  A shorter news cycle?

Whatever the reason, the mainstream media sources are rushing to publish questionable information and conclusions.  I suppose there are a few investors out there who made a decision based upon what George Soros was doing.  Some sources said this was evidence that “the big one” was at hand.

And of course, the government is no help.  The required filings are well-intentioned, but do not provide even the most basic information required to discern the actual position of these managers.

I always try to make a constructive suggestion.  Requiring the actual holdings — strike prices, both long and short — would allow readers of the filing to evaluate the position as of the stated date.  That is the best the government can do.  Trying to put a number on the position value is a moving target that changes with time and price.  Leaving out the short positions turns the exercise into a joke.

Misleading information is worse than nothing at all!

Most Commented On: Dollar Weakness and Stock Market Strength

It has now been five years since Jeff has been writing 'A Dash.' Before he left for vacation, he mentioned several of his regular themes that are of continuing importance for investors. We took a few of these ideas and found some interesting articles that you might have missed the first time, or might enjoy reading again. This first update consists of one of his most commented on articles. Whether remarkably insightful or just plain controversial, this post has garnered some of the most vocal attention from Jeff's readers.

Dollar Weakness and Stock Market Strength — the Data (November 14, 2009)

Anyone who is following the market on a daily basis is paying
attention to the strength of the dollar versus other currencies.  It
can be confusing.  While a strong dollar might seem like a good thing,
there is actually an inverse market relationship with stocks — at
least in recent times.

I described the basics of the dollar debate here:  Understanding the Debate on the Dollar.  There are many opinions — prescriptive, political, and descriptive.  The next step is to consider some data.

Dollar Strength and U.S. Equity Market Returns

my last article I highlighted the inverse relationship between a strong
dollar and the stock market.  This chart provides a dramatic
illustration of the relationship for calendar 2009.

Dollar one
Sometimes a chart tells the story in a glance, but I always like to look at data as well.  I used the Fed's measure of the dollar versus a broad basket of currencies, weighted by trade.  The series started in 1995.

The correlation for this year is an almost perfect negative relationship:   -.96.

A negative relationship sometimes stands out even more if you invert one of the scales.  Here is that chart.

Dollar two

The data are the same, but the tight relationship is especially clear using the inverted scale.

much for what we know.  An obvious question for investors in stocks is
what would happen if the dollar gets stronger.  Perhaps we should look
at more data.

A Longer Perspective

Some believe that
a strong dollar is consistent with economic growth and a strong stock
market.  To examine this concept, we need to consider more data.

Dollar three

we look at the full data series, we see long periods where dollar
strength was consistent with a rising stock market.  In fact, the
correlation between the two series is a positive value:  .35.

For an objective analyst of data, the chart suggests three conclusions:

  1. The relationship changes over time;
  2. The last year is quite atypical;
  3. There are other important variables at work.

The Carry Trade

The prevailing hypothesis about the
dollar/stock relationship is the "carry trade."  Supposedly there are
investors/companies/speculators who are able to borrow in dollars at a
very low interest rate.  This makes the dollar the "funding currency." 
Most outspoken on this subject is Nouriel Roubini, who sees disaster ahead.  [Full disclosure — I am a contributor at Nouriel's valuable site.  I respect and appreciate his encouragement of alternative viewpoints.]

borrowers do not then do the obvious — lend in another currency,
hedging the difference with a forex trade.  Instead they supposedly use
the borrowed funds to engage in a variety of speculative trades —
emerging markets, real estate, and, of course, US stocks.

Following this logic, there is now a speculative bubble of major proportions.  This seems implausible.

skepticism comes partly from the lack of any real data supporting this
viewpoint, and partly from the daily trading.  If you really had this
trade on, would you really take it off if the dollar moved by 0.4%? 
The carry trade hypothesis does not explain the closely calibrated
trade between the dollar and stocks, but we see this every day.

now, let us stick to conclusions supported by data.  There is a strong
inverse relationship.  If you are a trader, you had better pay
attention.  If you are a long-term investor, you need to understand the
long-term relationship.

Meanwhile, I invite reader comments,
citations for new data, and fresh hypotheses.  I will pursue this for
at least one more article, considering various causal models.