Graham's NCAV Strategy

In a previous post, we said that Benjamin Graham is considered the "father" of quantitative stock screening and investing. We also showed how his simplest screen, which attempts to "buy the cheap, high quality stuff", passes the test of time. Today we're going a step further.

Graham liked to find discrepancies between a stock’s price and its value and would buy  undervalued stocks, holding them until they became fully valued. In his book “The Intelligent Investor", Graham describes a stock selection technique that identifies stocks that are trading at a deep discount to a calculated value termed the Net Current Asset Value or NCAV.

In calculating NCAV, Graham only considered Current Assets (cash, cash equivalents, accounts receivable, inventories) and subtracted Total Liabilities. The result is the NCAV which, divided by the total number of shares, gives the NCAV/share. This value would be considered by Graham to be a fair value for a suffering stock.

You might think he would buy at this price, but no. In most cases, Graham only bought stocks that were trading under two-thirds or 66% of their NCAV. Experience has shown that it is safe to buy stocks up to 120% of NCAV and still make a profit.

Stocks trading at such a deep discount are few and far between, and have usually been beaten down by a combination of bad news and emotional reactions from the investing public. These stocks were Graham’s bread and butter. He repeatedly insisted that the time to buy stocks was when everyone else was selling and the time to sell was when everyone else was buying. 

Using Portfolio 123, we are going to test  the NCAV stock screener, using the interpretation seen also at Old School Value, which we believe is the closest to what Graham would use nowadays:

 

 No Chinese stocks

No financial or ADR or OTC stocks

Volume must be rising

NCAV/Mkt Cap > 1

Close > 50 cents.

And here's the first iteration:

NCAV all stocks, rebalanced quarterly.

But that means sometimes investing in 200 companies at the same time. 

What happens if we invest in 10 stocks only?

Results get reduced and there are periods of underperformance relative to the S&P.

Not all that great, considering the risks inherent in the strategy.

We are still rebalacing every 3 months though. What if we rebalance 

once per year? After all, NCAV stocks take time to develop.

NCAV with 10 stocks, yearly rebalance. Looking a lot better!

NCAV with 30 stocks, yearly rebalance, and we're back above the 15% p.a.

stated by Graham.

But the drawdown is huge. Is there any way to filter these NCAV stocks better?

I have tried to add additional filters such as:

a) high current ratio 

b) low earnings volatility

c) low accrural buildup

d) low debt/equity

...

But none of these permutations came close to the results of Graham's original screener. Why did the "quality" filters not enhance the results? Just remember what we're looking at: cigar butt companies that have serious problems, and that are selling for less than their core asset value. Graham himself said that companies will not get pushed below NCAV unless they are burning money, and have structural problems.

Essentially what we are betting on, when investing in Net Current Asset Value stocks, is that they will be bought out, or salvaged, for approximately their NCAV value per share. This is one drawback of the above backtest: it does not include any "sell" rules. The first and most obvious "sell" rule should be "sell if price reaches NCAV", because by definition there might not be much more upside. That's why the key is to buy stocks that are trading at a discount to NCAV. Graham himself wanted at least a 25-30% discount relative to NCAV, which he called "margin of safety". 

Again, it seems that the good old rules are robust and pass the test of time. Let's just remember why NCAV investing, a more aggressive version of value investing, works in the first place. It's based on human psychology. And so long as humans will make the market, there will be value to be found in these kinds of strategies.

Essentially investors tend to ignore cigar butts, largely depressed stocks, or stocks that aren't "cool to hold". Investors grativate towards glamour stocks, "story" stocks, "what's hot". However, the growth prospects baked into those stocks are already usually exhuberant. Vice versa, the "absence of growth" prospects are already baked into NCAV stocks...so it takes a very minor surprize to generate a substantial pop in price. 

Some of the most robust exploitable anomalies in the markets boil down to investor psychology.

The debate is open...what say you?