Remember how “say on pay” was supposed to put the lid on soaring executive pay? And just how has that turned out? According to a study conducted by ISS affiliates, (reported in Compliance Week ) the average compensation package for CEOs in the U.S. rose nearly 13 percent over the past year.  That study looked at companies in the Russell 3000 that had reported through April 13, 2015, a pool of about 1,200 companies.

This column from the New Yorker explores why, in the author’s view, say on pay hasn’t had the anticipated effect. The author observes that, in the past couple of decades, there have been a number of efforts to “change the mores of the executive suite,” but all attempts have ultimately followed the law of unintended consequences. For example, he asserts, the attempt to align the interests of directors and management with those of shareholders has made equity-based compensation a larger component of compensation packages. As the stock market has skyrocketed, the result has been substantial increases in the value of total comp. [And when there are declines, the notion of repricing and/or replenishing resurfaces.] Another attempt to impose good corporate practices, the column suggests, was to replace boards that historically were composed principally of “corporate insiders, family members and cronies of the C.E.O.” with largely  (purportedly, see this news brief) independent directors, so that now “C.E.O.s typically have less influence over how boards run. And S.E.C. reforms since the early nineteen-nineties have forced companies to be transparent about executive compensation. These reforms were all well-intentioned. But their effect on the general level of C.E.O. salaries has been approximately zero.” [In fact, some would argue that transparency has actually led to escalation of compensation as a result of benchmarking against publicly available data. See these news briefs.] Although executive compensation fell during the financial crisis, it “is now higher than it’s ever been.”  According to the column, “total compensation is up more than seven hundred per cent since the late seventies. There’s little doubt that the data for 2014, once compiled, will show that C.E.O. compensation has risen yet again.”

And here’s the most ironic part. While we know that, in a number of cases, shareholders have rejected pay packages, sometimes repeatedly, the percentage is very small: “Turns out, rather than balking at big pay packages, [shareholders] approve most of them by margins that would satisfy your average tinpot dictator. Last year, all but two per cent of compensation packages got majority approval, and seventy-four per cent of them received more than ninety per cent approval.”

In the author’s view, these reforms have been so ineffective in reining in executive pay because “they targeted the wrong things. People are justifiably indignant about cronyism and corruption in the executive suite, but these aren’t the main reasons that C.E.O. pay has soared. If they were, leaving salary decisions up to independent directors or shareholders would have made a greater difference. As it is, studies find that when companies hire outside C.E.O.s—people who have no relationship with the board—they get paid more than inside hires and more than their predecessors, too. Four years of say-on-pay have shown us that ordinary shareholders are pretty much as generous as boards are. And even companies with a single controlling shareholder, who ought to be able to dictate terms, don’t seem to pay their C.E.O.s any less than other companies.”

Rather, the author believes, there has been an “ideological shift. Just about everyone involved now assumes that talent is rarer than ever, and that only outsize rewards can lure suitable candidates and insure stellar performance.” However, the author asserts, that doesn’t necessarily prove to be the case; studies show that “if the company with the two-hundred-and-fiftieth-most-talented C.E.O. suddenly managed to hire the most talented C.E.O. its value would increase by a mere 0.016 per cent.”

He also contends that performance pay does not really work because the types of tasks that a CEO performs, such as deep analysis or creative problem solving, are typically not susceptible to performance incentives: “paying someone ten million dollars isn’t going to make that person more creative or smarter. One recent study… puts it bluntly: ‘Higher pay fails to promote better performance.’” In addition, the argument goes, performance is often tied to goals that CEOs don’t really control, like stock price (see this post and this news brief.)

[Sidebar: As discussed in this post, Professor William Lazonick has argued that stock-price-based performance targets may actually spur some executives to conduct stock buybacks to artificially boost stock prices. To address that issue, he recommends limiting the use of stock-based pay, subjecting incentive compensation to performance criteria that reflect investment in innovative capabilities and not stock price performance, and preventing executives from selling immediately upon exercise of options.]

What’s the bottom line? The author argues that “the failure of say-on-pay suggests that shareholders and boards genuinely believe that outsized C.E.O. remuneration holds the key to corporate success. Some of this can be put down to the powerful mystique of a few truly transformative C.E.O.s. But, more fundamentally, there’s little economic pressure to change: big as the amounts involved are, they tend to be dwarfed by today’s corporate profits. Big companies now have such gargantuan market caps that a small increase in performance is worth billions. So whether or not the people who sit on compensation committees can accurately predict C.E.O. performance—[one commentator] argues that they can’t—they’re happy to spend an extra five or ten million dollars in order to get the person they want. That means C.E.O. pay is likely to keep going in only one direction: up.“

A contrary view of the meaning of say-on-pay votes is presented in this news brief, discussing a Towers Watson survey showing that almost three-fourths of investors say executive pay is excessive, two-thirds think it is too heavily influenced by management, and only one-third agree that executive pay is closely linked to corporate strategy. Directors “overwhelmingly disagree on all these points,” with only one-fifth of directors concurring that executive pay is too high, one-quarter indicating that management has too much influence over pay and 70% believing that executive pay is closely tied to corporate strategy. According to the WSJ, “[t]he wide divergence in views suggests directors are wrong to infer from the typically high levels of approval in say-on-pay votes that investors on the whole support how they pay executives.” According to a Towers Watson representative, ” ‘[t]he say-on-pay vote is too much of a blunt instrument…. Investors may find a lot not to like in an executive pay package but vote for it anyhow. Things have to be pretty bad to say you don’t want to support the entire program.'”