Top-CEO Pay Isn’t Driven By Talent, New Study Says | Hacker News Reader

Top-CEO Pay Isn’t Driven By Talent, New Study Says

blogs.wsj.com

56 pointsanigbrowl55 comments

Yeah, but these are systematic problems. None of the super-rich CEOs deserve their pay, period. But the problem is a comparison. A CEO who feels like they are underpaid relative to other companies may feel undervalued merely by that comparison, regardless of being super wealthy.

The rise in CEO pay actually correlates partly with transparency. People thought reporting the pay would shame the waste, but it resulted in CEOs paid more reasonably feeling undervalued, and that drove their pay up.

There needs to be systematic rejection of this nonsense, like all companies agreeing together to pay more reasonably and otherwise pay employees reasonably.

The real best answer is probably a more progressive income tax (ideally with a Universal Basic Income funded by a negative income tax). Then CEOs can brag about how much more taxes they pay than others given how much higher their salary is, but we don't have to have so much waste and inequity.

I don't think you need to do much more than add more tax brackets and increase them. The top bracket really should be approaching like 90% for income in excess of $1M a year... IE. the tax brackets and rates should generate an income return that looks somewhat like a sigmoidal curve... if you've looked at biology this usually results (or is a result of) stable equilibria.

Also, capital gains gets taxed as income.

"Also, capital gains gets taxed as income."

That, is the most heart stopping sentence you can say to an American rich person. The tax structure is setup to tax the poor and middle class the hardest. If you are in a class that can do capital gains, 83b elections, double irish -- no taxes.

It's also not fair because you lose some of your investment's value to inflation.

So that 15% long term rate ends up closer to 30% if you include the fact that your dollar is worth less today than when you bought the stock.

I did some rough math on putting money into the stock market vs. inflation:

http://www.reddit.com/r/politics/comments/38b2u9/once_you_ar...

This is utter ballocks. Inflation affects everyone equally, there is no reason why people with enough unneeded income/capital to invest it should get tax breaks.

You know who really suffers from inflation? Minimum wage workers who's wages do not go up despite inflation.

Again, like I showed in my numbers, it's NOT a tax break. It's offsetting inflation. You pay very little inflation on your income this year, but you paid inflation "tax" on those investments from years ago you sold today.

The obvious solution to that would be to tax capital gains above inflation, using some government-standardized measure of inflation.

Yes, in an ideal world. But as you can see to responses, people still don't understand the point that I'm trying to make.

I would be completely fine with charging personal income rates on capital gains, as long as the capital gains are counted above inflation.

Giving you the benefit of the doubt this is really sloppy logic to argue what capital gains tax should be.

A 17 year doubling rate is a PA return of 4% if you're only expecting a 4% yearly return then yes shares aren't great, no shit. Also you can control your CGT liability by choosing when to sell.

On top of that you're doing calculations on stock market investment without looking at dividends? I mean given how cheap it would be for a billionaire to pay shills for this sort of thing, and how terribly damaging an increase in CGT would be for them.

Actually giving you the benefit of the doubt frankly you really SHOULD sell this sort of posted logic as a service. LAAS the next big HN thing perhaps? :D

Not my problem than the S&P 500 only returned 4% from 1998 to today. What do you want me to do, start counting from the bottom of the market? Nobody knows to buy at the bottom of the market because the bottom can only be known retrospectively.

Sure, if you invested in 2002 then your returns might be 2.5x today and the inflation lower, and the tax rate might be 20% (or whatever, I'd have to redo the numbers)

If you invested in 2000, however, you'd make 40% on your investment, lose money to inflation and THEN pay capital gains tax. You'd be getting taxed for losing money.

So I picked a more neutral date like 1998 where it wasn't a bubble yet, but not close after a crash. I didn't include dividends because I'd have to add those constantly as the divident yield from S&P 500 changes from month to month. It's around 2% (actually under 2% for the majority of the last 15 years), though, so that does change the numbers.

I would say then the inflation is around ~2%, dividends around 2%, stocks grew around 4%, so around 6% year-to-year nominal profit which you would pay 0.9% capital gains tax, and 4% real profit which is 22.5% capital gains tax (0.9/4 = 0.225)

Consider the following: A single individual with $200,000 salary with two personal exemptions would pay $45,000 in total taxes which is the same overall rate even though the marginal rate is 33%

So a rich individual who doesn't work, but only sits on $5,000,000 that yield him $300,000 a year (and that 5M loses $100,000 of purchasing power a year) would pay the same federal taxes as an individual who just works for his $200,000

So if you spend 10 years building a company and then sell it for 10 million, you think that you should only get to keep 1 million(ish)?

The only way to make a long term gains tax fair is to allow you to spread it out over the time it took you to make the gain. And even then it isn't really fair.

If you make a company and sell it in three years for 50 million, why is it fair that you only keep 5 million?

The only way to answer questions like this are by first asking our goals. Is our goal to let successful business leaders become ludicrously wealthy while externalizing much of the risk? Is our goal to have a system where the natural luck of the system creates lottery-style winners and losers and let that be? Or is our goal to have a maximally productive and healthy society?

>So if you spend 10 years building a company and then sell it for 10 million, you think that you should only get to keep 1 million(ish)?

Of course. By the time you're selling it, most of the actual value is being created by people beneath you.

Wait what? Two friends of mine sold a company for $10M last year, that was like them and 4 other employees, and the (technical) founders easily did more than half the work..

How much money did you make during that process?

I am not for 90% income taxes, but capital gains taxed as income is a good idea. I suggest starting with a rate between the current income and capital gains rate.

> like all companies agreeing together to pay more reasonably

I believe that's called wage fixing, and is illegal.

Or government regulation; in which case it is not illegal by definition.

As americans we are trained to forget that it is possible to have collective norms that are enforced by law. And indeed there are many who believe that all government regulations are ipso facto immoral; while conveniently forgetting the government granted monopolies and subsidies that make them wealthy. "Ayn Rand fuck yeah, but don't touch my soybean subsidies." 'merica!

> Or government regulation; in which case it is not illegal by definition.

Government regulations can easily be illegal. Happens all the time.

It doesn't have to be. Not for exceptionally high wages.

I'd rather see a system where a CEOs pay could be capped as a multiple of the lowest pay full time worker. ie. CEO payrises must coincide with base pay rates also increasing.

This is done at the Mondragon Corporation in Spain. Workers at different businesses/components of the firm vote to determine the ratio between the highest/lowest-paid workers. Last time I was looking at them (about eight years ago), 5:1 was the closest spread and 11:1 the widest. Most working class people recognize the importance of executives' contributions, but they were pretty confounded by the idea that anyone was sufficiently talented to merit 300:1 pay differential. So, the workers accept they may not have the most talented CEOs, but that they are also getting CEOs committed to the ideals of the firm.

I'd love to see that too, but instead every department of a company would become an "independent contractor" and the highest-level management would end up a tiny, 100 person company with salaries starting at 300k.

Or perhaps a large company of low wage workers overseen by 100 highly paid independent contractors hired by the board of directors.

Or you'll end up making the CEO more of a honorary or figurehead position where your only responsibility is to show up for board meetings once a month, and vesting the real decision making power (and big salary) with for example the COO/CFO/CTO or whatever.

Just make the lowest pay workers part time or contractors, then you'll be yielded a higher margin

Then just base the hourly CEO wage a multiple of the lowest hourly wage a sub sub sub contractor makes when working for the company. Including developping countries workers.

And suddenly every company would stop having full time workers, just a lot more contractors

that happens now already, so we need to fight with it anyway

Maybe government should tell companies that they need to treat "full time" workers differently from contractors and companies would gladly employ their workers full time.

"The rise in CEO pay actually correlates partly with transparency."

This seems reasonable, and I don't disbelieve you, but do you have any pointers to studies that back-up this conclusion?

Well, see https://www.fastcompany.com/1803069/social-medias-envy-effec...

Dan Ariely talks about this in his research, how important the comparison issue is. People feel more rewarded being the highest paid and undervalued being the lowest paid. As long as all numbers are above a comfortable base income, this comparative element overwhelms any actual significance of the absolute amount of money.

Paul Krugman has a somewhat related theory.

http://www.vox.com/2015/5/18/8619687/paul-krugman-ceo-pay

Switzerland publicly voted on a strict 12x salary limit for CEOs compared to junior employees. The proposal was rejected after 65% of voters opposed it.

http://www.bloomberg.com/news/articles/2013-11-24/swiss-vote...

The article also says that the same law (with public voting) would have had a much larger success rate in Germany where apparently three quarters of the population would vote in favor.

I wonder how such a public vote would turn out in the USA.

If it were up to the actual voters, I believe it would pass. But there is no way in hell any more than a small handful of American politicians would vote for this.

This has also been tried in the USSR. The economic success of such a measure is evident.

Yeah, there was definitely nothing else different in the USSR that might also make affect economic success. Apart from how they paid their CEOs, they were just like the USA.

How can the pay gap fluctuate so wildly in the last twenty years? I imagine compensation packages tied to the stock market drives some of that, but some of it seems pretty huge (376 -> 188 -> 345 -> 195 -> 303x.)

My question, out of curiosity, would be, if lowering compensation for top execs would not lower productivity (as the article predicts), why isn't it done then?

Think about it, CEO pay is determined by the company board. Those boards are made up of people who are often CEOs or former CEOs (look at the board of Apple for example 6 of 7 are CEO's or former CEOs) In other words you have a situation where a relatively small group of people are determining their own pay.

What do you think would happen to engineer pay if compensation were determined by a vote of a group of peers and you knew that the person who's pay you were voting on today might well be among the group of people determining your pay tomorrow. Personally I would always vote to increase pay by as large a number as I could possibly justify with a straight face.

That's a good point I hadn't thought about. So it's governed by game theory moreso than by market forces, i.e. even if candidate1 is not a great candidate, she's the best candidate available and we can get in our industry niche.

Perhaps a movement could be begun where companies use low ceo/worker multiplier as a marketing tool to highlight how pro-staff they are and by extension how pro-consumer they are.

I wonder why non-voting stock isn't more common, with a minimum number of people on the board being set if needed.

The question is, why didn't this happen before.

My 2 cents, the 80s is where automation truly started to take off, in the 90s, IT took over.

More work, less costs.

I've always wondered why instead of minimum wage laws they don't have a law defining the maximum difference % wise in total compensation between the highest paid and lowest paid worker at the company. That would at least align the CEO's incentives with that of their employees.

Would you advocate a similar system for equity between founder/CEO/execs and employees?

I find that people in tech are outraged by Big Corp CEO pay, but don't bat an eye when the founder of a successful startup walks away with $40M while the employees still live in rented condos and each get a fraction of a house down payment.

What do you mean by 'people in tech'? I don't think any of the engineers like it, VC probably do, but that's a very small subset.

Engineers accept it, every time someone accepts tiny equity at a small startup. I just don't see the outrage in our own industry.

It would also encourage CEOs to replace low-paid workers with expensive to maintain, buggy robots - sure, the company/shareholders might be somewhat worse off, but the CEOs wage could go up, and you can always misrepresent the costs and benefits, especially if there's an industry-wide trend in that direction. Or you could make contractors out of employees.

That basically fixes the CEO's wages at whatever that multiplier is * minimum wage because the lowest paid worker's wages probably won't increase over time. A large-enough company is always going to have entry-level employees and even if each individual's wages increase, the lowest paid worker is always going to be an employee who is brand new to the company in the lowest-paid position.

Hiring CEOs is a competitive marketplace like anything else and boards need to attract the best individuals in whatever way works best -- maybe the most fair way to compensate CEOs is through one-time, slow-vesting stock grants so that the CEO's "salary" they make each year is closely correlated to the value of the company (and therefore, to their success).

"A large-enough company is always going to have entry-level employees"

I bet they would outsource all such jobs, even more so than they do now. The typical big company doesn't employ cleaners or restaurant staff, they hire companies who do it for them.

I also bet that 'schools' would be created where what used to be new employees get trained for years, partly on the job, and get some compensation for it (a bit like the partner system in legal firms, but one where the not-yet partners technically work for a different corporal entity)

At the companies providing those low-paid services, there would be pressure to automate away the lowest paid jobs not because they can be done cheaper, but because they would allow management a pay rise.

It would be interesting to see what people would come up with, but I'm not sure I would find it an improvement over what we have now.

And that competitive market place may only exist because those in the set of CEOs artificially keep that set small by only shopping within their group.

I said _instead_ of minimum wage laws.

This article seems to be saying that a portion of CEO pay can't be attributed to free market competition for talent alone.

The Woodward Governor Company once did that. They limited their CEO to 10x the lowest paid employee of the company. That ended around 1980.

Because then the lowest paid worker in the company would be an executive, and all the other workers would become part-time contractors.

Four things spring immediately to mind.

1) This is research by a partisan think tank. That's not to impeach the source, only to encourage deep criticism about what the data is actually saying -- whether you agree with it or not. (In fact, I find that stories like this that I agree with I need to be especially careful of embedded spin because of the difficulty I have difficulty in seeing it.)

2) If we're all going to have robots working for us in 100 years, the effective ratio between the bottom workers and the top ones will be infinity -- that is, the entire idea of a happier future is machines doing all the grunt work and people just doing creative things that they find value in. This data may be a sign we are on our way there -- I'd be careful about destroying our future before we get there.

3) The measure here is being made with "comparably-paid workers" So the inference is that you can compare specialty doctors and CEOs, or stock brokers and CEOs. No matter how you justify it, I don't think such a comparison holds water. Are good stock brokers making 20x the money that bad stock brokers make? Perhaps. But does that mean that there would be a similar 20x range among CEOs? No, of course not.

4) I hate statistics-based social policy advocacy because there's always this underlying idea of the old "We take an aggregate number, make some broad generalizations about causality, then announce that the conclusion we had already reached before beginning our work is now evident." mentality. Carrots are poison? Sure, because 100% of the people who eat carrots are dead within 115 years. The only causal relationship I can come up with for CEO pay is the obvious one: highly-paid CEOs get that way because they manipulate the board into paying them that kind of money. This in itself is a skill. Creating large majorities of agreement among diverse stakeholders while maintaining a vision may also be a highly-paid skill among CEOS, and it may be related to their ability to manipulate boards, but even that is a bit of a stretch.

Beware facile arguments about complex topics.

I disagree. Negotiation is their talent, obviously.

wait- so what _is_ it driven by?

so I can min-max

Reply on news.ycombinator.com