Switch to Interactive Brokers

One year ago I did a major housekeeping change.

I finally opened an Interactive Brokers account  (I still have accounts with TDAmeritrade and OptionsHouse – now ETrade).

I don’t particularly like IB’s web interface or the desktop GUI. But one year in, it bothers me far less than it used to. I love the speed and the quality of execution. My other brokerage accounts, while offering better UIs, force me to bid in increments of 5c, and often take forever to fill comparable orders.

The big factor, of course, is cost. Instead of shelling $5+, most of my options cost me $1 or less. IB often gets me a price better than my limit price. I understand that it could pocket the difference, and I would never know. Actions like these build trust (Costco/Amazon playbook). In 2017, almost 90% of my trades were executed on IB.

It also pays me meaningful interest on my cash balance (0.92% currently for cash balances above $10k). This is higher than the interest at my bank.

As my portfolio moved from 16% cash at the start of 2017 to nearly 35% currently, the interest I earn underwrites all my commissions and then some.

Advertisements

Interesting Thread on Edge

Recently, I read two superb takes on whether retail investors have any meaningful edge.

Nate Tobik did a “Drake equation” to estimate that there are about 1000 pairs of eyes following listed stocks in the US. So really, our edge is nonexistent or smaller than we think.

Geoff Gannon generalized the discussion with a broader notion of “edge”.

He argues that unlike a casino, the stock market inherently has a positive edge. If you spend infinite time in a casino, you will go bankrupt. If you spend infinite time in the stock market, you will end up extremely wealthy.

So at the base level everyone in the market has an edge over those on the outside. This is a generic edge.

Then, there is the special edge.

Since buying certain kinds of stocks (high quality businesses, cheap stocks, and stocks rising in price) works better than buying other kinds of stocks (low quality businesses, expensive stocks, and stocks falling in price) an investor who systematically bets in order to maximize certain factors (like high quality, good value, and positive momentum) has an edge over both operators who systematically bet in order to maximize other factors (low quality, poor value, and negative momentum) and operators who don’t bet systematically.

And finally there is the stock-pickers edge, which Nate talks about.

Learning the Wrong Lessons

2017 was my first full year of “mindful investing”. As expected, I made many mistakes. These can be classified into two buckets.

Selling Winners Early

I sold WTW around $18. It soared to nearly $50 over the next few months.

I sold AAPL at an average price ~$130-140. It kept rising to $175.

Peter Lynch said, “Selling your winners and holding your losers is like cutting the flowers and watering the weeds.”

I should hold my winners longer.

Historically my worst mistakes have been “successful investments”.

I sold MSFT and EBIX for 80% and 300% gains, only to see both stocks double from there. And don’t even get me started about Priceline. I sold it only because I had made 10x on the stock. There was no fundamental reason for selling. It was agonizing to watch it soar 10x from that point, over the past 7-8 years.

Sucking my Thumb too Long

I dabbled with a few stocks, hoping to buy them at more attractive prices, only to watch them fly away. I watched FSLR go from $25 to $60, MSM from $70 to $95, FAST from $40 to $55, VFC from $50 to $75, WFC and NKE from $50 to $60+, and on and on.

So perhaps, the lesson is don’t be too picky. But is that the right lesson to draw?

I began the year with a certain view of the overall market: I expected it to return somewhere between -25% to +15%.

One can ask two questions: (i)  was the view correct, and (ii) was the strategy (of being patient with buying) correct, given the view?

In retrospect, the S&P clearly overshot the optimistic end of my expected range. The pendulum had swung more than I thought. My view was wrong.

However, given a view that is negatively biased, the strategy of waiting for more attractive prices was not necessarily wrong. Or at least, a single year (esp. one like 2017) may not be sufficient to draw that conclusion.

 

Oaktree Update

I looked at OAK over a year ago. At that time, I thought OAK was worth at least $38, and more likely close to $54 based on normalized earnings.

I added to the position earlier this year, when the price dipped to $40.

Over the past week or so it dipped back to the low $40s, after spending much of the year in the mid to upper 40s.

The overall narrative remains unchanged. It is a high-quality asset manager, focused on debt securities, which seeks to earn money the right way – by making outsized returns for its clients. With the stock market getting somewhat frothy, it remains poised to capitalize on any significant downturn or crash.

Valuation

I used the same template as the previous valuation. I used updated numbers from their Q3-2017 report for balance sheet items. For earnings, I used TTM numbers.

Screenshot from 2017-11-10 20-39-19

The year-to-year performance of OAK is quite lumpy. The IV increased from $38 to $66 based on TTM numbers. I believe that the average numbers are probably more reliable. They suggest that the IV remained roughly unchanged (increase from $54 to $57).

Based on these numbers, OAK is looking attractive at these levels.

Leucadia Update

In September 2016, I thought Leucadia was worth somewhere between $27-$32. At the  time, it was trading around $19, having bounced off of of lows in the summer. The company was still under-earning, although some of the winds had started to shift.

Earlier last month, the company put out an investor presentation (link), and I came across this well-reasoned article in SeekingAlpha, which pegs the intrinsic value around $33.

I updated my SOTP analysis, and come up with a fair value near $32.

Screenshot from 2017-11-10 07-37-11

The discount to intrinsic value has narrowed considerably from 50% ($19 versus $30 in 09/2016) to about 20% ($26/$32). The intrinsic value itself has improved marginally, mostly on the backs of improvements at National Beef and Berkadia.

The dividend has also been increased by 60% from 25c/share to 40c/share. At current prices, this implies a yield of over 1.5%.

I continue to hold.

Maintenance CapEx

Companies make money by selling stuff they make. To make stuff, they need to invest capital.

This invested capital (IC) often takes the form of PPE for traditional widget manufacturers. From a business standpoint, one can take a broader view of IC, and recategorize any expenditure that is borne in a particular period, but “pays off” over longer time periods as capital. Thus, R&D or subscriber acquisition costs (in sticky businesses) may be thought of as CapEx.

This might involve reinterpreting the CapEx that appears on the cash flow statement.

CapEx (CX from now on) has two components: maintenance capex (MCX), which a business has to reinvest just in order to stay in place, and growth capex (GCX), which is the part of CX that increases sales.

CX = MCX + GCX

MCX is a “cost”. GCX is an investment.

If you own a rental business for example, replacing a broken heater is MCX.

Buying a new rental property is GCX.

Warren Buffet’s owner’s earnings (OE) or Greenwald’s adjusted earnings try to account for this “cost” of doing business.

OE = NOPAT - MCX

Calculation

Calculating MCX is both art and science. It is not a number that is directly broken out in financial statements. However, we can develop some intuition on how to think about it broadly, and estimate it.

For a mature company with stable sales (no growth or decline), there is no GCX. Thus, all CX is MCX. For such stable businesses, MCX is approximately equal to the depreciation.

MCX = depreciation

Another popular method uses the following algorithm:

  • estimate the sales to capital ratio, SCR
  • compute the change in sales dS = S(t) – S(t-1)
  • GCX = dS/SCR
  • MCX = CX – GCX

Additional Notes

  • average SCR over several years; compare with industry averages
  • some CX may not produce sales growth immediately (ex. building a new theme park). This may involve projection, if the past doesn’t rhyme with the present.
  • if stable or meaningful SCRs cannot be extracted, then be wary of the number you produce.

References

  1. CS Investing: Calculating Capex
  2. Old School Value
  3. Seeking Wisdom
  4. Gurufocus

Cash Secured Puts: An Algorithm

Here’s a rough flowchart of my current CSP strategy.

Ideas - 31

I am still actively tweaking my process; so the following only represents a snapshot of my current thinking.

Search

The first step is to search a good candidate. Usually, I have about 20-30 stocks on my radar. These are stocks I am interested in owning. Some of these are companies I already own, but want to own more of.

In any case, I have a reasonable guess of the intrinsic value, IV, for this set of companies. Some of them may be trading below IV (price p < IV). If they trade at a sufficient discount, I will consider buying them outright. Usually, I like to buy stocks for at least a 20-25% discount to IV. This represents my margin of safety (MOS).

But what do I do, if the discount is not large enough? Say,

IV - MOS < p < IV.

I use this condition as an initial filter.

At any given time, depending on the overall market, 3-5 stocks may pass this filter. Two stocks, I am looking at, that pass this filter currently are MSM (IV ~ $85-90), and WFC (IV ~ $55-60). If MSM or WFC traded around $60-65 or $40-$45, respectively, I’d buy the stocks outright.

The second important filter is liquidity. When I am buying stocks, I usually don’t care about liquidity, because my holding periods are usually long. I don’t trade positions frequently. This freedom allows me to buy small-cap and foreign stocks.

With the CSP strategy, however, I need a fair bit of liquidity, since (option) holding periods (more on this later) are usually of the order of one month, and bid-ask spreads on illiquid names can be brutal.

This restricts the universe to somewhat large cap names, say market cap greater than $5B. This number is a guide, not a strict threshold. I like to see open interest for upcoming ATM options to be greater than a few hundred. Usually, both these facts are correlated.

WFC has a market cap over $250B. MSM is much smaller, ~$4B. The open interest for the $50 WFC put expiring on Oct 20 is about 5000. The comparable number for MSM is about 10. So WFC passes the second filter; MSM doesn’t.

The third filter is stock price. I like the stock price to be around $50 or below. This seems quite arbitrary, but is a self-imposed constraint because of position sizing and tactical considerations.

Right now, I like to restrict new positions that I ease into using CSPs to a maximum of 5% of my portfolio. At my current portfolio size, this means a total outlay of between $20-25K. Unlike many options traders, my goal – my ideal outcome – is to buy with a sufficient margin of safety. CSPs are a way to reduce the effective buy price below that level.

The main reason for the $50 stock price filter is that I might embark on long CSP campaigns. Sometimes, this means committing additional capital to the position if it moves deeper in the money.

I like to have 4 slugs in my revolver ($20K outlay/4 options ~ $5K/100 shares ~ $50 price threshold). This gives me added flexibility to adapt, adjust, and exploit volatility.

The final filter is price of the option. I target a minimum return of 20% IRR. If I commit a certain amount of cash to a new CSP, I seek annualized returns on that capital of 20% or more. I will discuss this filter in more detail in a separate post, because there is a fair amount of nuance.

If a stock has fallen in the recent past, its implied volatility is often high (say >30%). These are ideal candidates to chase, since the fall in stock price may have brought them close to value territory, making the pursuit all the more fruitful.

To recap, I search for the following characteristics:

  • attractive stock, but not cheap enough,  0.75 * IV < p < IV
  • liquid options market (generally large cap >$5B)
  • p <= $50, approximately
  • initial hurdle IRR > 20% (high implied volatility helpful – but check upcoming events)

Start Campaign

Okay, once we’ve figured out we want to pursue this strategy on an attractive candidate, we write or sell ATM puts on 1/4 or 1/2 of the intended position, with a duration of about a month. Why?

We sell at ATM options, because they have the thickest time value premiums. Ideally, we want to maximize these as we are going in. Later, we may be have to be content with ITM puts, which are less juicy, but we aim high to begin with!

How to size the initial position? This depends on how large the margin of safety is at the current price. If it is reasonably large, then I’d go with half. If it is not large enough, then I go with a quarter. This usually ties up $5-10K in the first round.

Finally, I usually try to sell options about one month out. You can get higher IRRs by using weekly options if they are available. This involves more frequent trading (potentially higher commissions), and careful upkeep. A frequency of one month is short enough to capture a fair chunk of the fastest decay in time value, and long enough to suit my temperament and schedule.

To summarize, we start a CSP campaign by,

  • writing ATM or slightly OTM puts,
  • on 1/4 – 1/2 the intended final position (depending on MOS),
  • with a duration of about a month.

Then we collect the premium and wait.

Quick Validation

Several things can happen. One outcome is that the underlying stock shoots up, and moves far away from the strike price of the puts we wrote. If that happens, it may be worthwhile to close the position prematurely. Suppose,

G = initial premium collected = maximum gain
T = initial duration of the put (in days)
t = current time to expiry 
g = gain on the position at current time (g < G)

Currently, I use a rough rule of thumb.

If the gain on the position exceeds 75% of the maximum possible gain, within the first half of the intended holding period, then I close position. Mathematically, this condition can be expressed as.

t < T/2 and g >= 0.75 G

The reason for not continuing to hold on to the position is that it frees up capital.

Thus, if the stock spikes, and then falls down again, you may be able to exploit the volatility by collecting twice on the same stock. In reality, things can work out even better. When the price of the security rises, its implied volatility falls. That, and the moneyness of the position make it cheap to buy it back. If the stock falls back, generally implied volatility increases, and you can juice out more time value from an ATM option.

Expiration

At expiry, you encounter one of three possibilities. Let’s dispose of the easy ones first.

If the stock prices exceeds the strike (p > S), you let the position expire. Depending on the price, you may consider reestablishing a new position at a later time.

If the stock prices hovers around the strike (p ~ S), or slightly below, you can roll out the position. This means you close the current put, and write a new put one month out. This lets you collect additional premium, or lower the effective buy price.

If the stock price has gone substantially below the original strike, things get interesting. This is where experience may be invaluable. I haven’t been doing CSPs systematically for long, but think that I have already gotten better at a few things.

In any case, you have several options; you could,

  1. take delivery – if you have reduced your effective buy price to a point where it is sufficiently below the intrinsic value, then it might be a good idea to take possession. Remember the overarching goal of the strategy is to buy good companies at a cheap price.
  2. roll out  – If you aren’t still happy with the effective buy price and have no new capital to commit, you can roll out the position. Note that the premium for ITM or deep ITM options may not be terribly attractive. Increasing the duration beyond one month might help add a little meat on the bone, but it will hurt your overall IRR. That said, it is a fine default strategy.
  3.  add position: if you have dry powder left (this is why I like four slugs in my revolver), you have more flexibility. There are multiple flavors, you could (i) keep the strike the same, (ii) lower it, or (iii) lower it and add extra duration. The first option (keep strike same) usually gives you the fattest premiums – it works best, when you think the stock has gotten really cheap and you want to lock in this cheapness. You fully expect the stock to rebound before the next expiry. The second or third options (lower strike/add duration) are worth considering, when you think the stock is headed even lower. You may take an IRR hit extending your position this way, but you are playing the long game. This way you reduce your effective buy price, and set yourself up to write future options that are less ITM (and collect fatter time premiums).
  4. use a wheel strategy: this involves some combination of the alternatives laid out above, and requires additional dry powder. You take delivery of the ITM options, write slightly OTM covered calls on these options, and use your dry powder to simultaneously write ATM or slighly OTM puts.

In future posts, I will highlight specific examples using this template.