Did Harvard Divest From Israel?

Globes reports:

In another blow to Israeli shares, the Harvard Management Company notified the US Securities and Exchange Commission (SEC) on Friday that it had sold all its holdings in Israeli companies during the second quarter of 2010. No reason for the sale was mentioned. The Harvard Management Company manages Harvard University's endowment.

Harvard Management Company stated in its 13-F Form that it sold 483,590 shares in Teva Pharmaceutical Industries Ltd. (Nasdaq: TEVA; TASE: TEVA) for $30.5 million; 52,360 shares in NICE Systems Ltd. (Nasdaq: NICE; TASE: NICE) for $1.67 million; 102,940 shares in Check Point Software Technologies Ltd. (Nasdaq: CHKP) for $3.6 million; 32,400 shares in Cellcom Israel Ltd. (NYSE:CEL; TASE:CEL) for $1.1 million, and 80,000 Partner Communications Ltd. (Nasdaq: PTNR; TASE: PTNR) shares for $1.8 million.

I'm not sure where Globes got the sales data. The 13-F form is a holdings report, not a transaction report. In other words, institutional investors use the 13-F to report their current holdings at the end of the quarter, not sales. But if you compare the first quarter and second quarter 13-Fs, the endowment management company owned the stocks in question at the end of the first quarter and no longer owned them at the end of the second quarter.

In one sense, Harvard did divest--they dumped all their Israeli stocks. But most people use divest in a more nuanced way; i.e., to intentionally sell and thereafter refrain from investing in stocks of a particular country for political reasons. So the interesting question is why Harvard sold the stocks at issue. Was it coincidence, a purely investment-driven decision, or a surrender to political activists opposed to Israel? Only the latter would count as divestment in my book.

I've written about divestment campaigns before, most notably the PC(USA)'s decision to divest from Israel:

Let's start with a basic question: Will the PC(USA)'s decision "work"? In other words, do divestment campaigns tend to achieve their proponent's goals? The clear answer from the empirical literature is "no."

A London Business School Institute of Finance and Accounting working paper called "The Effect Of Socially Activist Investment Policies On The Financial Markets: Evidence From The South African Boycott concluded:

"We find that the announcement of legislative/shareholder pressure of voluntary divestment from South Africa had little discernible effect either on the valuation of banks and corporations with South African operations or on the South African financial markets. There is weak evidence that institutional shareholdings increased when corporations divested. In sum, despite the public significance of the boycott and the multitude of divesting companies, financial markets seem to have perceived the boycott to be merely a 'sideshow.'"

Another paper, "The Stock Market Impact of Social Pressure: The South African Divestment Case," from the Quarterly Review of Economics and Finance in fact found:

"Using the South African divestment case, this study tests the hypothesis that social pressure affects stock returns. Both short-run (3-, 11-, and 77-day periods) and long-run (13-month periods) tests of stock returns surrounding U.S. corporate announcements of decisions to stay or leave South Africa were performed. Tests of the impact of institutional portfolio managers to divest stocks of U.S. firms staying in South Africa were also performed. Results indicate there was a negative wealth impact of social pressure: stock prices of firms announcing plans to stay in South Africa fared better relative to stock prices of firms announcing plans to leave."

In sum, divestment may make activists feel all warm and fuzzy, but the evidence is that (1) it has no significant effect on the target of the divestment campaign but (2) likely does harm the activists' portfolios. ...

Managers of pension plans are fiduciaries of the beneficiaries of those plans. When they pursue a social agenda nearly certain to result in poorer performance, they are disserving their beneficiaries. The activists at the PC(USA) may have gotten a warm and fuzzy feeling from taking a slap at Israel, but in doing so they injured Jewish-Christian relations, besmirched the one functioning democracy in the Middle East, and stabbed their own people in the back. All for the sake of a gesture that experience teaches will be fruitless.

One is left wondering whether Harvard went down the same path. Was this sale a political statement?

Update: As for the possibility that this was an investment decision rather than a divestment decision, a commenter at Atlas Shrugs opines that:

I disagree that it has anything to do with hatred on Israel. My source is the stock history of the companies cited over the past 6 months (presuming that Harvard has owned these stocks for this long): it lost money heavily on all of those companies except for Check Point, and even that was still a loss.

Therefore, it seems plausible that this was done to avoid taking a bath and makes 100 percent dollars and sense...with no hatred toward Israel shown here at all. SOURCE: Daily Finance 6 month stock history lookup.

This looks right to me. TEVA has been trending down steadily since April, for example, losing about 1/5th of its value. NICE has also been dropping over the period, losing about 1/6th of its value. CEL shed almost a third of its value in April and May. It could well be a pure investment decision with no political motivation. So some of the knee jerk hysteria may be misplaced.

Update: Joe Weisenthal notes that Harvard management sold its shares in an Israeli ETF and bought shares in a Turkish ETF, before remarking that:

Bear in mind that the infamous flotilla incident happened in Q2. Just saying.

What he doesn't tell you is that MSCI Israel is down 14% YTD, MSCI Turkey has basically broken even YTD. He also doesn't tell you that MSCI Turkey returned 44.74% over the last year, while MSCI Israel returned only 14.08% over that period.

If Harvard divested Israeli stocks for political reasons, I am perfectly prepared to criticize them, just as I did the Presbyterians. But the more I learn about this matter, the more it looks like a pure investment decision.

Update: The israel Matzav blog reports that there are two non-divestment explanations for Harvard's sale floating around:

One is that with Israel's admission into the OECD, we are no longer an emerging market, and therefore the emerging market section of the portfolio had to be reshuffled.

The other is that the Israeli stocks in Harvard's portfolio had performed poorly of late.

I actually find the first explanation more plausible than the second. University endowments are long-term investments and would not be likely to be reshuffled solely on the basis of a quarter or two of poor performance. On the other hand, the OECD admission is a recent, verifiable event.

The idea that University endowments are long term investors is probably wrong. Harvard Management Company outsources portfolio management to a number of institutional investor management companies. And while I haven't found any data on the average turnover rate for Harvard's managers, we know that most institutional investors have pretty high churn rates. A Mercer study, for example, found that:

Some active equity fund managers have higher portfolio turnover rates than they themselves claim, a new study finds.  Nearly two-thirds of institutional investor-focused investment strategies exceeded their expected turnover from June 2006 through June 2009.  Of these strategies, the turnover was on average 26 percent higher than anticipated, with some strategies reporting turnover between 150 and 200 percent more than expected. 

The new report, “Investment Horizons – Do Managers Do What They Say?” demonstrates that investment managers themselves underestimate turnover and often do not live up to their stated claims when it comes to the holding periods for the stocks in their portfolio.  The study, conducted by Mercer and funded by the not-for-profit IRRC Institute examines the investment horizon of active equity investment managers, comparing various strategies, different geographies and styles between June 2006 and June 2009....

Within the entire sample of 991 strategies, the average annual turnover of the sample is 72 percent, with some 20 percent of strategies having turnover of more than 100 percent.

Harvard speaks:

The Management Company's most recent SEC filing details changes in holdings, as is routine, but no change in policy. The University has not divested from Israel. Israel was moved from the MSCI, our benchmark in emerging markets, to the EAFE index in May due to its successful growth.

Our emerging markets holdings were rebalanced accordingly. We have holdings in developed markets, including Israel, through outside managers in commingled accounts and indexes, which are not reported in the filing in question.