The strategy in one paragraph

Graham defined a company's net current asset value (NCAV) as its current assets minus all liabilities and preferred stock, giving no value at all to factories, property or goodwill. When a stock trades below that figure, the buyer is getting the long-term assets and any future earnings for nothing. Graham preferred to pay no more than two-thirds of NCAV, and he bought these stocks as a group rather than one at a time. See how to calculate NCAV for the formula.

What Graham said about his own results

In The Intelligent Investor, Graham wrote that buying diversified groups of stocks below their net current assets produced satisfactory results in his experience for more than 30 years, roughly 1923 to 1957, with the exception of the severe market of 1930 to 1932. He ran the strategy at Graham-Newman, the investment firm where Warren Buffett worked in the 1950s.

Buffett used similar bargains in his early partnerships and later called the approach "cigar butt" investing in his 1989 letter to Berkshire Hathaway shareholders: a discarded cigar butt might have one free puff left in it. In the same letter he explained why he came to prefer buying wonderful companies at fair prices.

What the studies found

Henry Oppenheimer (1986)

"Ben Graham's Net Current Asset Values: A Performance Update," Financial Analysts Journal, November–December 1986

What was tested: US stocks at or below two-thirds of NCAV, bought each December 31 and held for a year, from 1970 to 1983. Portfolios held between 18 and 89 stocks.

  • A mean return of 29.4% a year, against 11.5% a year for the NYSE-AMEX index.
  • The compound (geometric) annual return was about 28.2%.
  • The results were uneven: from the end of 1970 to the end of 1973 the net-nets returned 0.6% a year, behind the index's 4.6%.
  • Net-nets that were losing money did slightly better than profitable ones.

Bildersee, Cheh and Zutshi (1993)

"The performance of Japanese common stocks in relation to their net current asset values," Japan and the World Economy, 1993

What was tested: Japanese stocks from 1975 to 1988. Too few companies met Graham's strict rule, so the authors used a looser test: any stock with positive NCAV relative to market value.

  • Market-adjusted returns of around 1% a month for the portfolio, roughly 13% a year above the market.
  • One of the first tests of the idea outside the United States.

James Montier (2008)

"Graham's Net-Nets: Outdated or Outstanding?", Société Générale, September 2008; reprinted in Montier's Value Investing: Tools and Techniques for Intelligent Investment (Wiley, 2009)

What was tested: A global basket of developed-market stocks below two-thirds of NCAV, 1985 to 2007, in US dollars.

  • More than 35% a year, against 17% a year for an equally weighted universe of stocks.
  • About 5% of individual net-nets fell 90% or more in a single year, compared with about 2% of stocks in the broader market.
  • Despite that, the portfolio had only three losing years over the period, against six for the market.
  • At the time of writing, Montier found around 175 net-nets worldwide, more than half of them in Japan.

Xiao and Arnold (2008)

"Testing Benjamin Graham's Net Current Asset Value Strategy in London"

What was tested: Stocks on the London Stock Exchange from 1981 to 2005 with NCAV at least 1.5 times market value, which is Graham's two-thirds rule.

  • Significantly positive market-adjusted returns, up to 19.7% a year, over five-year holding periods.
  • The premium remained after allowing for the small-company effect, and neither the CAPM nor the Fama-French three-factor model explained it.

Carlisle, Mohanty and Oxman (2010)

"Ben Graham's Net Nets: Seventy-Five Years Old and Outperforming"

What was tested: An update of Oppenheimer's US test using the same two-thirds-of-NCAV rule, from 1983 to 2008.

  • A mean monthly return of 2.55%, ahead of both the NYSE-AMEX index and a small-company index over the period.

Tobias Carlisle has also written about net-nets at length in Deep Value (Wiley, 2014) and, with Wesley Gray, Quantitative Value (Wiley, 2012).

How to read these numbers

The evidence is consistent, but the headline figures flatter what an investor would earn:

  • Averages vs compounding. Several studies report arithmetic mean returns, which are higher than the compound growth an investor actually experiences. Oppenheimer's 29.4% mean compares with a 28.2% compound return; for more volatile results the gap is larger.
  • Tiny, illiquid stocks. Net-nets are mostly micro-caps. Wide bid-ask spreads, commissions and taxes reduce real-world returns, and the strategy can only hold a limited amount of money.
  • Long dry spells. Even in Oppenheimer's strong sample, net-nets trailed the market for three years in a row.
  • Individual failures. As Montier found, single net-nets blow up more often than average stocks. Results came from holding many of them.
  • The past is not a promise. These are historical backtests, not guarantees of future returns.

Putting the strategy to work today

The studies share a recipe: a strict price rule (usually two-thirds of NCAV), a diversified basket, and regular rebalancing. To follow it you need a reliable list of net-nets, which is harder than it sounds once you look beyond one country. Our guide to finding net-net stocks covers how to build that list and a 10-point checklist for each candidate. If you want to go stricter than NCAV, see NCAV vs net-net working capital.

Build your own net-net portfolio

EveryNetNet lists every stock trading at or below NCAV in 50+ countries, updated daily. Start free with 50 US net-nets.

See Today's Net-Nets