> I've got no stake in this. I don't really care if buybacks are good or bad. But the author makes a point that buybacks, being so tax effective, make companies much more likely to spend their money on stock buyback rather than rainy day funds and investing in research/development/employees.
Can you point me to where they make that argument? I don't see it. They do sort of hint at it, here:
> To understand the magnitude of this shift, we analyzed financial data from 232 companies in the S.&P. 500 Index that were publicly listed in 1981, before the rule, and were still public through 2016. We found that from 1981 to 1983, these companies spent 4.3 percent of profits on buybacks. In comparison, from 2014 to 2016, these same companies spent 59 percent of their profits buying back their own stock. Dividends absorbed just under half of profits in both periods.
This data looks like it's trying to make you believe that these companies are allocating more capital to shareholder remuneration than they otherwise would. But it's not actually saying that. Dividends and buybacks are what you do with profits. If you re-invest your profits, they're not profits anymore, they're costs, so they aren't accounted as profit.
The point that you are making (that afaik, the article doesn't explicitly make) is a good one (if true): That buybacks shift the capital preference curve towards returning money to shareholders. If you wanted to prove that, you wouldn't look at the share of profits that go to buybacks, because all that would show you is that companies are preferring buybacks over dividends. Not that they are preferring buybacks over re-investment. Thinking briefly about it, you'd probably want to look at changes in revenue / capex, or changes in net-income to capex over time and correlate them with share of profits devoted to buybacks. AFAIK, the authors have not done this, and certainly haven't done it in this article.
EDIT: In a paper written by the authors, they do sort of do this, and it doesn't really show much:
> By decade, for 1984-1993, 1994-2003, and 2004-2013, total distributions to shareholders of these 248 companies
were 79 percent, 79 percent, and 84 percent respectively, with the proportion of net income devoted to buybacks
rising from 25 percent to 37 percent to 47 percent. High total payout ratios among major U.S corporations, therefore,
are not new, but over the past decade buybacks have predominated in distributions to shareholders.
Note: previously in the article they establish that preference for buybacks is very low in 1984, and goes up dramatically through to present (2013). So, 1984 is representative of a 'low buyback' time.
So, the payout ratio from net income went from 79 to 84 percent. That's not totally trivial, but it certainly isn't "save the economy" levels of relevant. They basically acknowledge this: "High total payout ratios among major U.S corporations, therefore, are not new, but over the past decade buybacks have predominated in distributions to shareholders.". But make no real attempt to reconcile this with their point. And notably, they make no attempt to control for other factors here. That 5 percent bump may be caused by higher margins (e.g. in tech) or any number of other economic factors. Being extremely generous, the data is suggestive of a slight preference shift for returning capital to shareholders over re-investment in the business. However, to actually conclude that you'd need to do something much more rigorous than this. And to further conclude that this preference shift has negative effects on the economy, you'd need to do a lot more than this.
Paper: https://www.brookings.edu/wp-content/uploads/2016/06/lazonic...