On Ben Graham's Net Nets

This is my first post, so please be gentle with me.

I thought I would put down some of my thoughts on some deep value situations: “Net Nets". Hopefully it could become a discussion point for anyone interested or who has experience in investing in net-nets.

Benjamin Graham first espoused his net current assets value (NCAV) rule for security selection in the 1934 edition of Security Analysis. For more info please see the excellent Stockopedia screen on this under the Bargain Stocks tab: Benjamin Graham NCAV Bargain Screen.

The investment rationale is summed up neatly by Carlisle et al [2010] "Graham proposed that investors purchase stocks trading at a discount to NCAV because the NCAV represented “a rough measure of liquidating value" and “there can be no sound reason for a stock's selling continuously below its liquidating value" (Graham and Dodd [1934]). According to Graham, it meant the stock was “too cheap, and therefore offered an attractive medium for purchase." In layman's terms if you were to liquidate (sell) the company: cash on the balance sheet is worth £1 in £1, inventories are marked down slightly and fixed assets are marked down more – eg who wants your 3 year old office PC even at written down book value? Once all the assets are added up and the debt and pension obligations taken away there is a positive value which is in excess of the stockmarket value (aka market Cap) of the company. You really are buying a £1 at a discount.

Graham applied his NCAV rule in the operations of his investment company, Graham - Newman Corporation, through the period 1930 to 1956. He reported that stocks selected on the basis of the rule earned, on average, around 20 per cent per year (Oppenheimer [1986]).

In the intervening years there been a number of academic studies of net asset value investing, including Oppenheimer [1986] and Carlisle, Mohanty and Oxman [2010]. All find that the returns to NCAV investing exceed those of passive investing, and by some margin. Carlisle et al conducted a study in assessing Net Current Asset (aka Net-Nets) investing between December 83 and December 2008. They found the "mean monthly return was 2.55%. This compares to the mainland to return the NYSE AMEX and small firm indices of 0.85% and 1.24% respectively".

It would appear therefore there is real value to hunting out these situations. In fact when I say real value, think about the compounding effect of 25 years at 2.55% per month. The portfolio would rise by 30% over 1 year, double in 2 years and rise by 700x in 25 years (pre stamp duty, etc but at this rate of return frankly who cares)….. Personally I might stop after 15 years, having made 50x my initial investment, as there is only so much money one needs. Also if you really did compound at 30% p.a. you would get too big for the companies you are investing in. So lets back off a bit from the Carlisle et al mean monthly returns and I suggest Ben Graham's 20% pa seems a more likely outcome.

So how do we ferret them out ?

Well, Google is a wonderful resource in this respect. Just type in "Stock Screener" and a plethora of screens jump out at you. Some are paid for, some are free. Here are some of my favourites: Stockopedia (of course!), Zacks (free), Old School Value (paid for), FT.com (Free!?! http://markets.ft.com/screener/customscreen.asp), FINVIZ (free: http://finviz.com/screener.ashx) Google Finance (free), Yahoo Finance (free, but basic). I will let you find your own favourites. There are amazing tools available now that weren't around in Benjamin Graham's time. Also given the amount of resource that is free, your main cost will be your time! No need for Reuters of Bloomberg costing at least $1000 per month.

What stands out is how in this information age with algos and bots engaged in flash trading, under the auspices of "smart beta" or "scientific active" strategies, how there is any alpha left for the rest of us. Why hasn't this just been arbitraged away ? If you dont believe me just read this overview from a well known trend following investment management house: " XXXX was founded in XXXX on the principle that financial markets are not efficient, and that an empirical scientific approach to investing can provide a long term edge." It goes on to say the company employs nearly 150 "data scientists and software engineers". That is some weapons-grade analytical hardware! Given this is just one firm amongst thousands, with automated rule-based trading strategies it remains a mystery why these NCAV situations haven't been arbed-out. Or does it ?

I guess the answer is that NCAV situations tend to be small in nature with Mkt Caps under £100m (I realise this is a huge generalisation), under researched, probably illiquid with large bid offer spreads and likely to trade infrequently such that the returns to the situations are not linear, rather they do nothing for months then 'Boom' rise spectacularly because something changes. When you have a CTA (Commodities Trading Advisor) Fund with client money of lets say $1bn, leveraged 5x = $5bn AUM. It would be like forcing the entire water mass contained in the Hoover Dam through the eye of a needle if they wanted to trade in and out of these small cap situations. So they don't. They trade in more liquid situations instead, thus leaving the playing field for others.

Looking at the Carlisle study, which was based on the NYSE, AMEX and NASDAQ exchanges, it documents 1362 NCAV situations over a 25 year period between 1984 and 2008. On average that's 56 per year, of which over 75% were listed on the Nasdaq. My guess is a large number of these were young biotech or tech companies which raised cash on IPO, with an idea and probably no revenues and then afterwards the market cap traded below the value of cash on the balance sheet. Only 245 out of 1362, or roughly 10 per year were from the larger NYSE. Interestingly the returns from the NYSE listed net nets were lower by c80bps per month (0.8% per month) when compared to the Nasdaq listed net nets.

Looking at the data it would appear to me that you need a portfolio of net nets and DYOR is imperative. There is just too much risk in 1 or 2 situations. Why do I say this ?
Lets take a hypothetical example:

Company: Alpha plc
Cash on Balance Sheet: £10m
Fixed Assets £1m (Graham typically would have discounted these at c 50%, but lets make the maths easy and leave them at 100% of value)
Net Debt: £0
Alpha plc Net Current Assets= £10m – 0m debt = £10m
Alpha plc Net Asset Value = £10m cash + £1m fixed assets - £0m debt = £11m

Assume 1m shares in issue at a market price of 70p = Mkt Cap £7m
Therefore its trading at 0.7x NCAV

So far so good. Masses of margin of safety.

But what if we assume the company is a tech start up and hell bent on producing the next, but better, Stockopedia and are spending £3m per year to develop the mother of all websites? Well after 12 months assuming they still haven't finished it and have no paying customers, the cash has depleted by £3m to £7m and intangibles have probably risen on the B/S. Where is our margin of safety now ? Especially if the database was found to be flawed and will take another £2m to rectify. Suddenly we are down to £5m cash and assuming an unchanged market cap of £7m, it is trading at a premium to NCAV. Ouch !

Therefore the prudent thing to do is have a portfolio of these situations. Don't rely on one or two. The difficulty as we saw earlier is that there are really so few of them.
I wonder though if it is possible to weed out the dubious situations (eg Naibu?) and concentrate on the more solid net nets eg GENCOR (GENC US). Maybe however this subjective intervention reduces return? Maybe one for a budding PhD student to explore ?

Hopefully some food for thought and I would really appreciate any comments or better (!) any net nets that people have identified where they think they can make a sensible valuation case.

More on GENCOR and Naibu soon.

Have a great 2015 !