統計的有意

Let’s be clear about what must stop: we should never conclude there is ‘no difference’ or ‘no association’ just because a P value is larger than a threshold such as 0.05 or, equivalently, because a confidence interval includes zero. Neither should we conclude that two studies conflict because one had a statistically significant result and the other did not. These errors waste research efforts and misinform policy decisions. …

These and similar errors are widespread. Surveys of hundreds of articles have found that statistically non-significant results are interpreted as indicating ‘no difference’ or ‘no effect’ in around half (see ‘Wrong interpretations’ and Supplementary Information).

See also Supplementary Information, American Statistical Association, NPR

Go-Shops Revisited

The Original Study, which examined deals announced in 2006-07, reported that a higher bid emerged during the go-shop period 12.5% of the time (6 instances out of 48 go-shop deals). Using a new database of M&A transactions over the past nine years, we find that the jump rate in the 2010-2018 timeframe was 5.6% (6 out of 108 go-shops), declining to 2.5% (1 out of 40) in the period 2015-2018. The last successful go-shop in our sample occurred approximately three years ago, in January 2016, when II-VI Inc. successfully jumped GaAs Labs’ offer for ANADIGICS, Inc. during a 25-day go-shop period.

As one of us concluded in the Original Study, “go-shop provisions can be a better mousetrap’ in deal structuring – a `win-win’ for both buyer and seller.”  However, over the ensuing decade, a broader set of transactional planners distorted the go-shop technology in ways that achieve their clients’ objectives but no longer satisfy broader corporate law objectives of promoting allocational efficiency in the M&A marketplace. (footnote omitted)

via Harvard, SSRN

Fernan Restrepo, Judicial Deference, Procedural Protections, and Deal Outcomes in Freezeout Transactions: Evidence from the Effect of MFW

This work next explores the effect of MFW on the gains of the target shareholders, as measured by the premium over market prices, the cumulative abnormal returns around the announcement of the transaction, and the change from the buyer’s first offer to the final offer. This part of the analysis considers two hypotheses. The first hypothesis predicts that the gains of the target should be higher after MFW because shareholder voting acts as a check against negligent or captured boards, and even if boards discharged appropriately their fiduciary duties, the target shareholders can still use the threat of a veto to push acquirers to raise their offer (Subramanian, 2005, 15; Edelman and Thomas, 2015, 468; Jiang, Li, and Mei, 2016). As a result, the fact that MFW effectively incentivized MOM conditions should lead to an upward pressure on deal prices. The second hypothesis suggests, in contrast, that the target gains should remain similar after MFW because freezoeuts were already subject to significant scrutiny before 2013, and judicial scrutiny appears to be an effective substitute for procedural protections (Subramanian, 2007; Restrepo, 2013; Restrepo and Subramanian, 2015). As discussed in Section 4, the results generally support this hypothesis.

Yakov Amihud, Markus M. Schmid & Steven Davidoff Solomon, Do Staggered Boards Affect Firm Value?, — Iowa Law Review — (forthcoming 2017)

We attempt to resolve conflicting empirical results in studies on the wealth effects of staggered boards by addressing issues of endogeneity, omitted variable bias and functional form. In a sample of up to 2,961 firms from 1990 to 2013 we find that additional variables provide significant explanatory power for the (negative) wealth effects of staggered firms found in prior studies, and their inclusion makes the effect of a staggered board on firm value insignificant. When we control for endogeneity by the instrumental variable method, we find again that the staggered board has no significant effect on firm value. Our results suggest caution about legal solutions which advocate wholesale adoption or repeal of the staggered board and instead evidence an individualized firm approach, and provide some measure of skepticism for law-related corporate governance proposals generally.

もう二度と計量経済学に触れない学生への計量経済学の教え方

Marc Bellemare writes:

私が理解したところでは、この論文は、予測やデータ生成過程の適切なモデル化を教えるべきではない、とは言ってない。ただ、それを教えるのは、因果関係の推定や研究デザインの根底をなす、相対的に単純で、難易度が低く、それでいて同程度に有用な考えを教えた後にするべきだ、と言っているのである。

私の見たところでは、多くの学部生は計量経済学の別の講座を取ることは決してない。学部生を教える教師としての我々の仕事は、責任ある市民を育てることであって、研究者を育てることではない。私が思うに、単なる相関に基いてなされた因果関係の主張にだまされないことを教えるのが、学生にとってよりためになると思う。私にとってそれは、回帰にどの検定方法を当てはめるかを知るよりも遥かに重要な、批判的思考法の構成要素である。

via himaginary氏

Lawrence Glosten, Suresh Nallareddy & Yuan Zou, ETF Trading and Informational Efficiency of Underlying Securities

Using a large cross-section of ETF holdings data from January 2004 to December 2013, we document that an increase in ETF trading is accompanied by an increase in price informational efficiency of the underlying stocks, as reflected in the increase in the relation between stock returns and earnings news. The effect of ETF trading on information efficiency should be conditional on the information environment and the degree of capital market competition. Consistent with expectations, when we conduct the information efficiency tests within different segments of the market, we find significant and improved informational efficiency among small firms (firms with market capitalization below the NYSE 50th percentile), stocks with low analyst following (firms with analyst following below the 75th percentile), and stocks with imperfectly competitive equity markets (number of shareholders below the 75th percentile). In contrast, we are unable to document such improvement for big firms, stocks with high analyst following, and for stocks with perfectly competitive equity markets.

Influencing Control: Jawboning in Risk Arbitrage

In an “activist risk arbitrage,” a shareholder attempts to change the course of an announced M&A deal through public campaigns, and profits from improved terms. Compared to conventional (passive) risk arbitrageurs, activists target deals susceptible to managerial conflicts of interest (e.g., going-private and “friendly” deals) and deals with lower announcement premiums. Their presence increases the sensitivity of deal completion to market signals. While they block a significant proportion of planned deals, activist arbitrageurs only modestly decrease the probability that the targets will eventually be acquired (including by a third party). Finally, the strategy yields significantly higher returns than passive arbitrage.

via Harvard