> However, looking at it as a percentage of GDP is flawed and leads to an understatement of the effect. We need to look at it after adjusting for inflation only, since all that matters is how big these companies are in the absolute after applying the fixed threshold cutoff of 500. If you're normalizing by GDP, you're not comparing apples with apples when doing a temporal comparison.

There are plenty of historical companies with higher market caps than today's companies in 2021 dollars. And again, if you compare the largest companies in 2021 with the largest companies in 1965 (using 2021 dollars), you'd find that their market caps are only ~2.5-3x the size (but yet their CEOs are paid 15x more relative to the average employee). But I think this misses the mark because a bigger company isn't necessarily harder to run. Nor is it clear that the CEO is making dramatically harder decisions. In fact, it seems equally plausible that the larger the company, the less of an effect the CEO has.

> That's right, and I outlined some of the reasons for that, such as commodification of domestic lower skilled labor due to globalization and technology.

This argument doesn't hold water, because the average employee salary (the one we're comparing the average CEO salary to) hasn't dramatically dropped in terms of purchasing power. In other words, your argument is that the average salary has gotten lower and therefore CEO pay appears higher - this isn't the case.

> This hinges on how we define a "good job".

If CEO pay was correlated to performance, we'd find that CEOs whose companies did better were better compensated. But we don't find this - we find that CEO pay is most correlated to the stock market.

> This zero-sum mindset is just not accurate. If Tim Cook gets paid $500 million and his decision making leads to $50 billion in extra benefit for the org compared to what a second-best CEO could have brought in, then the org has gained $49.5 billion in excess value. Your mindset is that the org (and the people inside it) have lost ~$500 million without considering the unique value add that makes it worthwhile and win-win for everyone. If the board didn't think that Tim Cook was adding unique value above his compensation, then they would give him a pay cut or hired someone cheaper.

This isn't true, because as noted above there isn't evidence that better-performing CEOs are paid better. Tim Cook isn't being paid his salary because the board believes he's adding such-and-such amount in value, but instead because CEO compensation is not related to the value added.

However, you didn't answer my final question: is there a level of inequality that you would consider unjustifiable?

  "In fact, it seems equally plausible that the larger the company, the less of an effect the CEO has."

This is easily shown false by simply comparing Apple with a small accounting firm. The max effect a CEO in the latter can have is on the order of $1 million, which is about 4 or 5 orders of magnitude less than the negative consequences to Apple if its CEO sucks. For a real world example see what happened when Ballmer was replaced by Satya Nadela. Or see what happened to Enron when a criminal CEO was steering the ship.

  "your argument is that the average salary has gotten lower"

That is not my argument. My argument is that the increasing supply of low skilled labor has meant that wages have not gone up alongside productivity growth, which contributes to the wage gap with CEOs. That's a subtle but important difference.

  "If CEO pay was correlated to performance, we'd find that CEOs whose companies did better were better compensated."

That's exactly what we find. CEO comp in the form of stock and options are usually tied to milestones or stock performance. Look at Musk's arrangement. The stock price is often used as a proxy for performance because it aligns incentives and is harder to game than explicit metrics.

  "is there a level of inequality that you would consider unjustifiable"

Not at all. If someone invents nuclear fusion in their basement and delivers $50 trillion worth of value to humanity, I really could not care if they keep a $5 trillion slice of that. They've gifted $45 trillion onto the species and I am extremely grateful to them for doing that.

> This is easily shown false by simply comparing Apple with a small accounting firm. The max effect a CEO in the latter can have is on the order of $1 million, which is about 4 or 5 orders of magnitude less than the negative consequences to Apple if its CEO sucks. For a real world example see what happened when Ballmer was replaced by Satya Nadela. Or see what happened to Enron when a criminal CEO was steering the ship.

All you're really saying is that bigger companies have more to lose, not that CEOs of bigger companies have larger effects. It could be easily argued that the CEO of a smaller company is able to more directly affect the business - fewer intermediaries making decisions, less oversight etc. As for your real-world examples, you can find countless counterexamples of companies who replaced "good" CEOs and performed just as well or perhaps better. In big organizations the factors that lead to success extend far beyond the CEO.

> That is not my argument. My argument is that the increasing supply of low skilled labor has meant that wages have not gone up alongside productivity growth, which contributes to the wage gap with CEOs. That's a subtle but important difference.

Just to be clear here, I'm talking about domestic workers. So arguments about globalization etc. don't apply.

> That's exactly what we find. CEO comp in the form of stock and options are usually tied to milestones or stock performance. Look at Musk's arrangement. The stock price is often used as a proxy for performance because it aligns incentives and is harder to game than explicit metrics.

Exactly my point.

1) If CEO compensation was based on performance, then their compensation would rise and fall based on performance relative to their peers. Instead, CEO compensation rises and falls relative to the overall strength of the stock market. Just graph executive compensation vs value of the S&P or Dow and this is easy to see. (Put another way - were 100% of CEOs doing a terrible job in Spring 2020 or Fall 2008?)

2) Since CEOs are compensated in stock (or, historically, stock options), their compensation has nothing to do with the "value" they bring and everything to do with the value of the stock market.

> Not at all. If someone invents nuclear fusion in their basement and delivers $50 trillion worth of value to humanity, I really could not care if they keep a $5 trillion slice of that. They've gifted $45 trillion onto the species and I am extremely grateful to them for doing that.

Executives of the British and Dutch East India Companies were probably compensated at rates tens of thousands of times greater than that of their average "worker". Do you consider that justifiable? What about companies that use (and used) slave labor?

  "not that CEOs of bigger companies have larger effects."

I am arguing this. Enron and Microsoft are examples. What small company CEO is capable of directly blowing up billions of dollars? Or with Nadela, pulling off a turnaround worth multiple billions? These are larger effects that are inaccessible elsewhere.

Your point seems to be that a small company CEO has a bigger impact relative to the size of the company. I agree with you on that. If a company has one employee - the CEO - that will be true by definition. But it's a trivial point and not one that is relevant because a one person operation who increases revenue from $200,000 to $400,000 shouldn't be compensated more than a large company CEO that drives an unexpected 5 percent growth in $100bn of revenue. The latter has much more value add and people that are thought to be able to do it will therefore be paid much more.

  "So arguments about globalization etc. don't apply."

It is highly relevant. It's why US manufacturing collapsed, because China does it cheaper and better. The average Joe now has to compete with an increasingly global workforce who have a lower cost of living and no minimum wage laws.

You are correct that domestic jobs like Uber drivers don't face international competition. But that's missing the point that low skilled labor is a fungible commodity. If all the factory workers get laid off, they start competing with Uber drivers for those same domestic jobs which increases the labor supply for a shrinking number of jobs and depresses wage growth.

  "Instead, CEO compensation rises and falls relative to the overall strength of the stock market."

This is a great point but I will take issue with your claim that it has "nothing" to do with performance. The reason stocks have positive beta in general is because the macroeconomy largely drives revenues for most sectors. Apple will legitimately make more money if the overall economy is healthy.

And not all stocks have positive beta. Biotech for example.

But I actually agree that beta should be stripped out when determining CEO pay, not because the market moves don't correlate with company performance (it does due to the macro confounder), but since market moves are merely good luck and not attributable to the skill of the CEO.

It doesn't happen which I see as a failure of boards.

At the same time, I can sort of see the validity in keeping it rather simple. The shareholders just want the stock to go up, and adding complexities that might disalign incentives could lead to bad second order effects such as a CEO pursuing an unnecessary hedging strategy (consider a CEO of an airliner who would be tempted to over hedge oil prices) in order to secure outperformance and therefore compensation in a downturn. But on balance, I do support the idea of stripping beta out.

  "Do you consider that justifiable? What about companies that use (and used) slave labor?"

I believe I addressed that in my initial post. If they're making money not due to legitimate value add, then no I do not think they're entitled to keep their wealth. Whether that's via rent seeking, pollution, slave labor.

There's obvious grey areas where the devil is in the details and a binary answer isn't appropriate. Airliners that contribute to pollution, consumer goods companies that might have some child labor in the supply chain, etc. I can't give a one sized fits all answer to these cases. But if we pick specific examples like SpaceX or the hypothetical person that invents fusion, then I want them to keep all of what they earn.

> I am arguing this. Enron and Microsoft are examples. What small company CEO is capable of directly blowing up billions of dollars? Or with Nadela, pulling off a turnaround worth multiple billions? These are larger effects that are inaccessible elsewhere.

But Enron is the absolute extreme case, not the typical case - and even there, there were large numbers of individuals involved in Enron's scandals so you can't just point to the CEO. Where's the evidence that Nadela was the primary factor in Microsoft's performance in the last 10 years? Was Steve Jobs a bad CEO because he was run out of a (then-failing) Apple? The point I'm trying to make here is that I don't think there's any evidence that CEOs at most big companies are "adding value" on the scale required to make your argument work - that CEOs deserve their extremely high pay because they provide extremely high value. (Do CEOs in the United Kingdom, Japan, or Germany provide substantially less value?). I think it seems much more reasonable that large company performance has to do with a huge number of factors, of which CEO performance is but a small one, and that CEO pay has more to do with the tax/regulatory regime as well as political and cultural attitudes.

> It is highly relevant. It's why US manufacturing collapsed, because China does it cheaper and better. The average Joe now has to compete with an increasingly global workforce who have a lower cost of living and no minimum wage laws.

It's a little bit relevant. The average Joe now has to compete with an increasingly global workforce, but that may actually increase the average typical salary of domestic workers as much as decrease it, because jobs that can be done more cheaply are not done in the United States.

> This is a great point but I will take issue with your claim that it has "nothing" to do with performance.

Sorry, "nothing" was a little hyperbolic.

> I believe I addressed that in my initial post. If they're making money not due to legitimate value add, then no I do not think they're entitled to keep their wealth. Whether that's via rent seeking, pollution, slave labor.

I'm glad we agree then inequality is political and not some emergent economic property :). You simply happen to buy into the current political justification for inequality ("people who add more value earn more money") but not myriad past ones ("the nobility and the clergy provide important functions to a harmonious society" or "property rights are inviolable").