Geoff Gannon September 29, 2013

How Did Mohnish Pabrai Not Make Money in Japanese Net-Nets?

This is a serious question. I’m probably not qualified to answer it, because I have a poor understanding of Pabrai’s investment approach.

Here is the blog post that stumped me:

The Pabrai funds invested in a basket of Japanese (net-nets) starting October 2010. Mohnish has exited all the positions with a realized gain of 2.2% including dividends, or 1.4% annualized.

The notes go on to give examples of some stocks Pabrai owned:

Examples include Hibiya Engineering and Ryoyo Electro. Both were trading below NCAV at the time of his investment, generating profits as well as positive and consistent cash flows. Managements of both companies were also repurchasing shares. Hibiya ended up just a little bit profitable and Ryoyo turned out be a -15% loss.

I can’t explain Pabrai’s experience with Japanese net-nets. But I can tell you a little about my own. We have two records of it. One is personal (my own account where I actually bought net-nets and made money). The other is public (a paid report I put out on March 21st, 2011). There is a third record you should check out. Go to Nate Tobik’s Oddball Stocks and read all his Japanese net-net posts. These three experiences are more indicative of what individual American investors would have gotten out of their Japanese net-nets.

I won’t talk a lot about my private record, because it’s private. You can’t verify it. But it’s better than the public record I’m about to show you. Instead of picking 15 Japanese net-nets, I went with no more than 5 at a time. I ended up buying a total of 6. I started with 5 and then added one later to replace a stock (Sanjo Machine Works, which is on the list) that was bought out. Like always, I concentrated a little more than other investors might. In this case, that got me a better personal result than the more diversified group I’m about to show you.

So let’s talk about that group of 15 net-nets. Let’s talk about the public record.

I published a (paid) report on Japanese net-nets on March 21st, 2011. So we have a list of 15 Japanese net-nets we can look back on without the usual biases of a hypothetical backtest. This is an actual observation. We’re working off a dated PDF that went out to buyers.

Here are the returns (in Yen) of those 15 Japanese net-nets since March 21st, 2011.

Zaoh (9986): 84%

Fuji Electric Industry (6654): 23%

Mitsui Knowledge Industry (2665): 28%

ASICS Trading (9814): 82%

Sonton Foods (2898): 47% TAKEOVER

Nisshin Electronics Service (4713): 33%

Daito Gyorui (8044): 9%

Sanjo Machine Works (6437): 186% TAKEOVER

NJK (9748): 67%

Noda Screen (6790): 62% TAKEOVER

M.O. TEC (9961): 41% TAKEOVER

Yasuhara Chemical (4957): 1%

Techno Associe (8249): 71%

Kawasumi Laboratories (7703): 12%

Seiko PMC (4963): 83%

Now, the Yen has fallen 18% against the dollar since March 21st, 2011. So, we will factor that into our results. Taking the currency loss into consideration, here are the cumulative (since March 2011) dollar returns an American investor who acted on my net-net report might have gotten.

Minimum: (17%)

Maximum: 168%

Median: 26%

Mean: 35%

All 15 stocks had positive returns in Yen. Only 12 of 15 had positive results in dollars. Four of the 15 stocks were taken over.

Pabrai had a realized gain of 2.2%. To put this in perspective, 6 of the 15 net-nets in my report returned at least 40% since March 21st, 2011. Again, that’s in dollars. Returns over 40% (in dollars) were actually more common than returns under 10%. The group median – which should be the most telling number – was a 26% cumulative return. That’s in dollars and ignoring dividends.

Pabrai’s result is nowhere near that. I would love to take credit for being some sort of especially good Japanese net-net picker. But I know that’s not true. I think if you look at other value bloggers that also bought Japanese net-nets – you’ll see similar results.

I think individual American investors who bought Japanese net-nets in March 2011 and held them till today generally got a median result of 20% to 30% cumulative returns. They usually had very few losers. And they almost always had a couple takeovers.

Pabrai’s result in Japanese net-nets was awful. And it was clearly his fault. Now, he may – because he is managing a lot of money – have focused on big net-nets.

I always say that’s a bad idea. The bigger the net-net, the worse the business. I think you should focus on the smallest net-nets you can find. But if you’re managing a hedge fund, you can’t do that. Small and boring is best. Hedge funds can do boring. They can’t do small.

This is not meant to be a criticism of Pabrai. He has – over his career – gotten better annual results than I have. He’s obviously a better investor than I am. And yet he obviously did a bad job implementing his Japanese net-net strategy.

The real purpose of this post is to point out to the folks reading the notes to Pabrai’s annual meeting that individual value investors – including Americans who obviously lost a lot on the Yen decline – made a lot more money in Japanese net-nets than Pabrai did.

It would be a shame if people read Pabrai’s annual meeting notes and figured Japanese net-nets didn’t pan out. They did pan out for people reading and writing blogs like this one and Oddball Stocks. They didn’t pan out for Pabrai.

Look at the names listed above. Those are the 15 names – and the only 15 names – that were in my net-net report. I can also tell you that I put my money where my mouth was. So, whatever liquidity costs were – I’ve come out way ahead after all those real life issues of liquidity, currency, and broker fees.

I would also encourage everybody to check out Oddball Stocks. I think Nate’s experience in Japanese net-nets has been similar to mine. And very different from Pabrai’s.