Regular readers of Greenbackd know that I’m no fan of “the narrative,” which is the story an investor concocts to explain the various pieces of data the investor gathers about a potential investment. It’s something I’ve been thinking about a great deal recently as I grapple with the merits of an investment in Japanese net current asset value stocks. The two arguments for and against investing in such opportunities are as follows:
Fer it: Net current asset value stocks have performed remarkably well throughout the investing world and over time. In support of this argument I cite generally Graham’s experience, Oppenheimer’s Ben Graham’s Net Current Asset Values: A Performance Update paper, Testing Ben Graham’s Net Current Asset Value Strategy in London, a paper from the business school of the University of Salford in the UK, and, more specifically, Bildersee, Cheh and Zutshi’s The performance of Japanese common stocks in relation to their net current asset values, James Montier’s Graham’s net-nets: outdated or outstanding?, and Dylan Grice’s Are Japanese equities worth more dead than alive.
Agin it: Japan is a special case because it has weak shareholder rights and a culture that regards corporations as “social institutions with a duty to provide stable employment and consider the needs of employees and the community at large, not just shareholders.” In support of this argument I cite the recent experiences of activist investors in Japan, and Bildersee, Cheh and Zutshi’s The performance of Japanese common stocks in relation to their net current asset values (yes, it supports both sides of the argument). Further, the prospects for Japan’s economy are poor due to its large government debt and ageing population.
How to break the deadlock? Montier provides a roadmap in his excellent Behavioural Investing:
We appear to use stories to help us reach decisions. In the ‘rational’ view of the world we observe the evidence, we then weigh the evidence, and finally we come to our decision. Of course, in the rational view we all collect the evidence in a well-behaved unbiased fashion. … Usually we are prone to only look for the information that happens to agree with us (confirmatory bias), etc.
However, the real world of behaviour is a long way from the rational viewpoint, and not just in the realm of information gathering. The second stage of the rational decision is weighing the evidence. However, as the diagram below shows, a more commonly encountered approach is to construct a narrative to explain the evidence that has been gathered (the story model of thinking).
Hastie and Pennington (2000) are the leading advocates of the story view (also known as explanation-based decision-making). The central hypothesis of the explanation-based view is that the decision maker constructs a summary story of the evidence and then uses this story, rather than the original raw evidence, to make their final decision.
…
All too often investors are sucked into plausible sounding story. Indeed, underlying some of the most noted bubbles in history are kernels of truth.
As to the last point, arguably, the converse is also true. Investors have missed some great returns because the ugly stories about companies or markets were so compelling.
There are several points that are not contentious about an investment in Japan. The data suggests to me and to everyone else that there are a large number of net current asset value bargains available there. The contention is whether these net current asset value stocks will perform as they have in other countries, or whether they are destined to remain net current asset value bargains, the classic “value traps.” My own penchant for value investing, and quantitative value investing in particular, makes this a reasonably simple matter to resolve. I am going to invest in Japanese net current asset value stocks. Here are the bases for my reasoning:
- I believe that value investing works. I believe that this is the case because it appeals to me as a matter of logic. I also believe that the data supports this position (see Ben Graham’s Net Current Asset Values: A Performance Update or Lakonishok, Shleifer, and Vishny’s Contrarian Investment, Extrapolation and Risk). Where a stock trades at a significant discount to its value, I am going to take a position.
- I believe that Graham’s net current asset value works. In support of this proposition I cite the papers listed in the “Fer it” argument above.
- I believe that simple quantitative models consistently outperform expert judgements. In support of this proposition generally I cite James Montier’s Painting By Numbers: An Ode To Quant. Where the data looks favorable to me, I am going to take a position, and I’m going to ignore the qualitative factors.
- I believe that value is a good predictor of returns at a market level. In support I cite the Dimson, Marsh and Staunton research. I am not dissuaded from investing in a country simply because its growth prospects are low. Value is the signal predictor of returns.
The arguments militating against investing in Japan sound to me like the arguments militating against any investment in a NCAV stock, which is to say that they are arguments rooted in the narrative. I’ve never taken a position in a NCAV stock that had a good story attached to it. They have always looked ugly from an earnings or narrative perspective (otherwise, they’d be trading at a higher price). As far as I can tell, this situation is no different, other than the fact that it is in a different country and the country has economic problems (which I would ignore in the usual case anyway). While the research specific to NCAV stocks in Japan is not as compelling as I would like it to be, I always bear in mind the lessons of Taleb’s “naive empiricist,” which is to say that the data are useful only up to a point.
This is not to say that I have any great conviction about Japan or Japanese net current asset value stocks. Far from it. I fully expect, as I always do when taking a position in any stock, to be wrong and have the situation follow the narrative. Fortunately, the decision is out of my hands. I’m going to follow my simple quantitative model – the Graham net current asset value strategy – and take some positions in Japanese net nets. The rest is for the goddess Fortuna.
When evaluating a single company as a net-net you need some basis for belief that the company could be liquidated. You further need a belief that they are not going to squander their net assets. This would likely need to apply across a basket also.
In terms of simply models (as suggested by Montier) it may be worth looking at how many (%) companies in Japan have been liquidated versus the US. It would also be worth looking at whether the net assets of net nets have been increasing or decreasing over time.
Net nets work because there is a fundamental valuation basis for them to work. That’s underpinned by the ability for investors to realize those net assets through some sort of process. That would be undermined by the evaporation of those net assets or by a cultural/ societal / legal refusal to liquidate.
(xposted from MOA)
LikeLike
I don’t know if you are right or wrong. But I do know one thing! Your discussion was first class end an enjoyment to read. You made my Sunday, even if I cannot buy stocks in Japanese companies. I live in Europe and my broker does not trade Japanese stocks and I do not think these kind of stocks have ADRs on NYSE.
LikeLike
How do you plan on screening for these companies?
LikeLike
The central question seems to be–does the prospect of liquidation need to be a plausible future course of events for the NCAV model to work–i.e., provide decent returns and protection of principal?
I would say, yes, it does. If companies does not have to worry about activists taking control and liquidating, then it is in management’s interest to continue operating with utter disregard for the stock’s low price. It’s hard for me to make sense of something like Klarman’s liquidation valuation, if there is little to no prospect of that occurring in individual cases.
It seems almost as absurd as saying–hey, if Buffett ran this insurance company, it would be worth X. If this company were liquidated, it would be worth Y. Both are similarly speculative.
Now, perhaps there are a handful of cases in Japan in which activists–domestic or otherwise–agitated for liquidation and were successful. Just a few cases would likely be enough to justify using a liquidation valuation, and commend the NCAV approach.
I don’t know of any, but that’s likely a mark of my ignorance more than anything. I know very little about Japanese market operations.
LikeLike
I really enjoyed reading both this post and your series on Japanese Net Nets. I think your analysis is first class. The ‘narrative trap’, as you have well documented, is simply making up a story about some evidence (i.e. a company’s earnings history) and making a decision based on that story. This is very dangerous because, by doing so, you often ignore critical evidence that conflicts with the story. Of course, it is one thing to recognize a problem and quite another to solve the problem. As an investor, the way I try to get around this problem is to predict, at a minimum, an upside, central case and downside scenario for a business, and attach probabilities to each of these scenarios. I think doing so helps prevent me from coming up with a story about whether I should buy the business. It does so because I am forcing myself to predict more than one future for the business. Thus, (adjusting Montier’s diagram) by predicting scenarios, my process will be to:
Gather Evidence > Weigh and Evaluate Evidence (Predict Scenarios) > Decide.
I discuss this process in greater detail on my investing blog the http://thefallibleinvestor.com
LikeLike