Tea Party, Hypocritical Parasites?

[cemper93]

I can't see how one can believe it's not. If most of the world's richest people are living in a country, but this country's general standard of living is only average, that's a sure sign of social inequity.

Dude its not average. It's not anywhere close to average. It's one of the highest, by some measurements the highest. Actually by average(mean), it is pretty much the highest by any adjustment. By median, it is always very high and often the highest.

I meant "average" both in the casual meaning (referring to an "average American", not the mathematical mean) and, of course, in comparison to other countries of the first world.

[Dark567]

I meant "average" both in the casual meaning (referring to an "average American", not the mathematical mean) and, of course, in comparison to other countries of the first world.

Sure then use median and compare to other first world countries, still at or near the top.

[Arrian]

In the discussion of a wage gap, isn't comparing the cost of labor to capital measuring the wrong things? The "owner" of capital in most large companies isn't the CEO or the company president, but the company itself as an incorporated entity. As I understand it if you're taking a system-wide "cost-of-labor" metric you're measuring both the high and low ends of the spectrum, the janitor and the CEO in the same measurement. To say that cost-of-labor and capital growth are in step would mean that the CEOs aren't getting their pay by taking it disproportionately from the incorporation, but it doesn't address whether or not the range of highest and lowest remunerations (of all types) is widening.

Remember that a large portion of CEO compensation comes from stock, ownership shares in the company, that is; capital. Indeed, the vast majority of the top 1%'s income is now from wages but from ownership interest in companies, which are returns to capital. The firm owns the capital, so what's left over after paying for labor is the return to capital, and the profit is what's left over from the cost of supporting capital.

The vast majority of people who are making millions of dollars a year are not getting that as a paycheck, they're making that off investments, which is another way of saying, they're getting the returns to capital. So yes, the wage gap between the top and bottom is in large part due to capital verses labor.

I can't see how one can believe it's not. If most of the world's richest people are living in a country, but this country's general standard of living is only average, that's a sure sign of social inequity.

Compare the definition of mean to median and think about what you said: "OTOH, 34% of world's billionaires live in the USA, which is a distant first place. If the median income is only "somewhere in the top 10", that's a huge contrast."

[cemper93]

Compare the definition of mean to median and think about what you said: "OTOH, 34% of world's billionaires live in the USA, which is a distant first place. If the median income is only "somewhere in the top 10", that's a huge contrast."

Actually, that difference in definitions is exactly what I was referring to. A country with lots of billionaires will always have a high average income because the famous one percent are significantly boosting it.

[maybeagnostic]

Compare the definition of mean to median and think about what you said: "OTOH, 34% of world's billionaires live in the USA, which is a distant first place. If the median income is only "somewhere in the top 10", that's a huge contrast."

Actually, that difference in definitions is exactly what I was referring to. A country with lots of billionaires will always have a high average income because the famous one percent are significantly boosting it.

That's not what median means. The top 1% do not change the median regardless of how much they actually make.

[cemper93]

Compare the definition of mean to median and think about what you said: "OTOH, 34% of world's billionaires live in the USA, which is a distant first place. If the median income is only "somewhere in the top 10", that's a huge contrast."

Actually, that difference in definitions is exactly what I was referring to. A country with lots of billionaires will always have a high average income because the famous one percent are significantly boosting it.

That's not what median means. The top 1% do not change the median regardless of how much they actually make.

Are you guys even reading what I write? I know that billionaires don't change the median (but the average) income, and that was precisely my point. I wanted to say that when the USA have the highest amount of billionaires in the world and therefore a high average income, it is a sign of social inequity that the median income is not the highest in the world.

[Dark567]

Are you guys even reading what I write? I know that billionaires don't change the median (but the average) income, and that was precisely my point. I wanted to say that when the USA have the highest amount of billionaires in the world and therefore a high average income, it is a sign of social inequity that the median income is not the highest in the world.

It doesn't work that way. The only reason the US has more billionaires than any of the countries above it, is because we are much much larger country(every country above the US, has a smaller population of NYC alone). You would need to look at per capita billionaires, which only has an effect on average income anyway, so would still be misleading.

[lutzj]

I can't see how one can believe it's not. If most of the world's richest people are living in a country, but this country's general standard of living is only average, that's a sure sign of social inequity.

Dude its not average. It's not anywhere close to average. It's one of the highest, by some measurements the highest. Actually by average(mean), it is pretty much the highest by any adjustment. By median, it is always very high and often the highest.

It's also the largest county in its income tier by one or two (edit: forgot Luxembourg etc, or three or four) orders of magnitude. You're going to see fewer billionaires in Norway because the size of the national and regional economy doesn't support as many billionaires, not because Norway is much poorer.

[Arrian]

Are you guys even reading what I write? I know that billionaires don't change the median (but the average) income, and that was precisely my point. I wanted to say that when the USA have the highest amount of billionaires in the world and therefore a high average income, it is a sign of social inequity that the median income is not the highest in the world.

But its median income IS among the highest in the world. It's awfully hard to get a country of 300 million people to beat a country of 500,000 people (or even 16 million) in every statistical category. It's entirely within the realm of possibility that a small, racially homogenous country might have some reasons for a slightly higher median income than a huge, diversely populated nation for entirely non-political reasons. Or, even have a higher median income due to politically negative reasons, like selective immigration rules. (See point 3.)

[Ghostbear]

If you look at the BLS's Employment Cost Index you'll get a better picture of what companies are paying for employees, a much more accurate view of compensation than simply looking at wages. Employment costs for all civilian workers are up 15.5% since 2005 and have about tripled sine 1981. Wage stagnation is not compensation stagnation.

I agree that compensation is more than wages, but you can't look at total costs, because that includes factors that aren't compensation, such as taxes and managerial costs. If you look here, you'll see that benefits make up about 30% of total compensation, with the rest being wages. Even if we assume that, when the wage stagnation started on those charts I gave, that all of compensation was represented by wages, this would indicate only a ~30% boost in compensation over the past 30 or so years. Compare that to an 80% boost in productivity, a 240% increase in income for the top 1% (almost certainly to be majority capital based), and (vs. 1990 instead of 1980) a 300% increase in CEO pay and 100% increase in corporate profits. And that is, again, assuming the best case scenario for labor compensation.

None of this is to say that the ratio of pay to labor and capital hasn't changed. The gains from increased productivity aren't necessarily being evenly distributed between labor and capital, but Ghostbear is arguing that all the gains from productivity increases are going to capital, and that's not the case.

I've been arguing that a disproportionate share is going to capital, not all.

Generally speaking, that is how it is in the technology sector. Fewer and fewer people are needed to generate the revenue. By this measurement, Facebook employees on the average are 10x more productive than Walmart employees. I'm almost certain that the average wage at Facebook isn't 10x higher than the average Walmart Employee.

Anyway, since the 80s are right at the point where "things change" according to your graph, that really makes me wonder about the effect of the technology sector on wages. If indeed, its because technology engineers / workers are say 10x more productive but only demand 2x or 3x the wages, then the graph begins to make sense.

Yeah, I wonder how much of this can be attributed to the fact that if you're Microsoft, the development costs for the newest Xbox are probably going to be relatively constant whether you end up selling 5 million or 500 million of them. Technology markets seem to have a lot of oligarchy-ness to them (natural or not- in many cases, e.g. processors, I'd assume it's a natural market trend), but unlike many other markets (such as retail), their employment needs don't rise very linearly with their market size. If you cut down on the market for labor, then you can get away with paying that labor less. The problem here is that the profits still stay up, and that most of that gets concentrated in the capital owners.

[Vaniver]

I've been arguing that a disproportionate share is going to capital, not all.

But, the proportionate share is determined by the price system. You can't just say "too much" is going to capital- how would you determine the proper share for capital?

[Ghostbear]

But, the proportionate share is determined by the price system. You can't just say "too much" is going to capital- how would you determine the proper share for capital?

If someone thinks the price system is broken (for me to have the position I do, it's implied that I do think it's broken), then saying that the current situation is determined by it is relatively pointless. Capital owners, such as CEOs, have a huge say in their own salary, both directly (by owning voting shares and actually using them) and indirectly (through serving on the boards of other companies and having an "scratch my back, I'll scratch yours" situation on raising everyone's pay). When one group gets to decide what the price system is, they're obviously going to favor themselves (and not out of some evil shadowy cabal or desire to keep the middle class down- it's just self interest). You should know from previous discussions that I don't ascribe any infallibility to the market or it's "guiding hand".

As for how I would determine it: I wouldn't. I'm not stupid enough to draw a line in the sand and say "above this, everything is OK, below, it is not". For a rough statement of what I'd expect for proper allocation, I'd say it's when compensation and wages rise at a similar rate to productivity, which the data I have provided (even if you include non-wage compensation) indicate it has not for about the past 30 years, and by a wide margin.

[lutzj]

If someone thinks the price system is broken (for me to have the position I do, it's implied that I do think it's broken), then saying that the current situation is determined by it is relatively pointless. Capital owners, such as CEOs, have a huge say in their own salary, both directly (by owning voting shares and actually using them) and indirectly (through serving on the boards of other companies and having an "scratch my back, I'll scratch yours" situation on raising everyone's pay).

The CEOs of the biggest corporations (i.e., those that can pay their CEOs million-dollar salaries) almost never own enough of a stake in the company they run to have any real influence. Based on their performance (mostly, measured as their ability to produce value for shareholders) CEOs are paid, and given bonuses, and have their bonuses withheld, and are hired and fired and headhunted just like many other employees. Most shareholders don't want to overpay the CEO, but when one increases the value of a billion-dollar company by 3%, it's easy to justify paying him $5 million every year.

CEOs usually have a lot of influence over their own compensation, but that has more to do with the importance of their jobs and their influence on the success of their companies than anything else.

[cemper93]

But its median income IS among the highest in the world. It's awfully hard to get a country of 300 million people to beat a country of 500,000 people (or even 16 million) in every statistical category. It's entirely within the realm of possibility that a small, racially homogenous country might have some reasons for a slightly higher median income than a huge, diversely populated nation for entirely non-political reasons. Or, even have a higher median income due to politically negative reasons, like selective immigration rules. (See point 3.)

Yes, I looked that up and you're right. Of course, this doesn't mean that income equality in the US isn't high (Gini 47%) and rising.

[Ghostbear]

The CEOs of the biggest corporations (i.e., those that can pay their CEOs million-dollar salaries) almost never own enough of a stake in the company they run to have any real influence. Based on their performance (mostly, measured as their ability to produce value for shareholders) CEOs are paid, and given bonuses, and have their bonuses withheld, and are hired and fired and headhunted just like many other employees. Most shareholders don't want to overpay the CEO, but when one increases the value of a billion-dollar company by 3%, it's easy to justify paying him $5 million every year.

I don't believe that follows reality: go here, then click on pay vs. performance. The CEO's of the S&P 500 ("biggest corporations) appear to have no substantial link to pay increases and company performance. Looking for other sources, it appears to also follow suit for CEOs in Massachusetts. And here's an older article from 2003 detailing CEO compensation in Indiana.

Bonuses for CEOs of Indiana's largest public companies jumped nearly 75 percent last year, even as the economy struggled and shareholder returns for all but four of the companies dropped.

While compensation for CEO pay and company performance is tenuous at best, I'm going to stick with what I said: CEO's have a large say in their own compensation. CEO's are usually the chair of the board of directors, and will often get to select it's members (here). Members of the board of directors usually run unopposed. CEOs (and other executive officers) will frequently serve on the board of directors of other companies*. The people determining a CEO's pay happen to be the people for whom they determine the pay of somewhere else! How can you possibly conclude that executive compensation is determined by their talents? Some of them (HP comes to mind) are frequently led by a bunch of no talent assclowns, where the company would probably be better off without any CEO at all, yet they'll frequently still receive higher salaries in a year than most people will make in a lifetime.

* And for some more anecdotal data: Paul Otellini (Intel) serves on Google's board. Eric Schmidt (Google) served on Apple's board. Intel's board of directors has 11 people, and includes one former CEO, two people who serve on three other boards, three other CEOs, and their own CEO. Walmart's board is similar: several former or retired CEOs, some still serving CEOs, and several people otherwise involved in the management of other companies. If you look at PepsiCo, you see it again: only one member who is not a current or former member of the executive leadership of a company. Similar trends exist for Microsoft, Bank of America, Yahoo, Verizon, ExxonMobil, DuPont, and.. Do I really need to keep going?

[KnightExemplar]

The CEOs of the biggest corporations (i.e., those that can pay their CEOs million-dollar salaries) almost never own enough of a stake in the company they run to have any real influence. Based on their performance (mostly, measured as their ability to produce value for shareholders) CEOs are paid, and given bonuses, and have their bonuses withheld, and are hired and fired and headhunted just like many other employees. Most shareholders don't want to overpay the CEO, but when one increases the value of a billion-dollar company by 3%, it's easy to justify paying him $5 million every year.

I don't believe that follows reality: go here, then click on pay vs. performance.

Are we looking at the same graph? Sure, there are outliers that prove your point, but for every one that I see, there's a Eric Schmidt who's taking home $1 in salary as the CEO. On pay vs performance, the "pay up as performance drops" (the upper-left hand side) is the minority of CEOs. The majority of CEOs improve the performance of their company without receiving any pay increases from 2008 to 2009. In fact, the majority of the CEOs on that chart are under the right side of "shareholders win". Unfortunately, the axis is on a different scale so perhaps the "squashed" nature of the graph is misleading...

Second, the definition of "Fair Pay" is an arbitrary calculation... and the graphs don't seem to have any adequate explanation on just what "Fair Pay" is. Generally speaking, companies are only compared to their peers: smaller companies are expected to grow much much faster than the S&P 500 for example, and Real Estate companies are expected to grow slower (but with less volatility) than tech companies. I have to question the authority of the "Fair Pay" numbers when they aren't really adjusted per industry.

EDIT: And third: if Fair Pay is partially based on the stock's price... then we already know that a stock's performance is not necessarily correlated to the company's profits, revenue, cash flow, or any other measurement of a company's performance. (A technical analyst would argue that its correlated to the psychology of the investors for example. Trend lines, "barriers", etc. etc) The stock's price apparently contributes to 2/3rds of the "fair pay" model... which IMO makes assumptions that I'm not comfortable with.

The CEO's of the S&P 500 ("biggest corporations) appear to have no substantial link to pay increases and company performance. Looking for other sources, it appears to also follow suit for CEOs in Massachusetts. And here's an older article from 2003 detailing CEO compensation in Indiana.

I admit that the Massachusetts graph shows a bit more on your point, but even then, the outliers on both sides make you wonder what is going on. Ex: Synta Pharmaceuticals Corp gained 5000% in revenue, but the CEO lost 15% on his salary. Perhaps they expected him to have an even higher revenue increase??

Unfortunately, you seem to be arguing that CEOs have control over their salary, while these graphs only note that their salary is not necessarily correlated to performance in a single year. You'll have to put forth data in support of your argument if you want to prove your point. (As opposed to what you're doing now: putting forth data in opposition to your opponent's point). Especially because there are very many questions that are raised by these graphs. (Do the salaries correlate to company performance from two years ago? Is there "lag" in determining a CEO's pay? Or does the CEO's salary correlate to projected performance? )

--------------

EDIT: For example: WHY do the charts focus so much on revenue? A company can make profits while its revenue drops: for example, if they downsize and lay off a fair number of workers, profits go up and the company can potentially become more sustainable despite a drop in revenue. Ultimately, the random nature of those graphs only tells me that a more complicated model than "pay vs revenue" needs to be considered. Clearly, we aren't getting the real picture from those graphs.

[Ghostbear]

Are we looking at the same graph? Sure, there are outliers that prove your point, but for every one that I see, there's a Eric Schmidt who's taking home $1 in salary as the CEO. On pay vs performance, the "pay up as performance drops" (the upper-left hand side) is the minority of CEOs. The majority of CEOs improve the performance of their company without receiving any pay increases from 2008 to 2009. In fact, the majority of the CEOs on that chart are under the right side of "shareholders win". Unfortunately, the axis is on a different scale so perhaps the "squashed" nature of the graph is misleading...

First, did you miss clicking on ""Pay vs. Performance"? That'd change your interpretation of the data quite a bit, I think (since you'd be looking at the wrong data).*

Ignore which quadrant of the graph they're in for a moment, and just look at how poorly correlated things are:

If there's any actual correlation between shareholder performance and CEO salary, I sure don't see it there. You'd expect a difference between a company that did 20% better and company that did 60% better, but there doesn't appear to be one (and if there was, it won't be very statistically correlated). You'd expect a much better relationship- maybe not strictly linear, but some sort of a trend to appear. There isn't one.

EDIT: Or taken another way: the chart shows that you have no real predictive power based off of one set of data. If I told you that one of the corporation's returns went up 30%, do you think you'd be able to guess an accurate increase in executive salary? If I told you that a CEO's salary changed, would you be able to guess what happened to the corporations returns? I sure wouldn't be able to.

* To be safe, here's a screencap of it:

Spoiler:

Second, the definition of "Fair Pay" is an arbitrary calculation... and the graphs don't seem to have any adequate explanation on just what "Fair Pay" is.

Which is why I didn't use the "Fair Pay" stuff at all in my argument.

Unfortunately, you seem to be arguing that CEOs have control over their salary, while these graphs only note that their salary is not necessarily correlated to performance in a single year. You'll have to put forth data in support of your argument if you want to prove your point. (As opposed to what you're doing now: putting forth data in opposition to your opponent's point). Especially because there are very many questions that are raised by these graphs. (Do the salaries correlate to company performance from two years ago? Is there "lag" in determining a CEO's pay? Or does the CEO's salary correlate to projected performance? )

And that is what the bottom half of my post covered! Requoted in a spoiler box to prevent possible confusion:

Spoiler:

While compensation for CEO pay and company performance is tenuous at best, I'm going to stick with what I said: CEO's have a large say in their own compensation. CEO's are usually the chair of the board of directors, and will often get to select it's members (here). Members of the board of directors usually run unopposed. CEOs (and other executive officers) will frequently serve on the board of directors of other companies*. The people determining a CEO's pay happen to be the people for whom they determine the pay of somewhere else! How can you possibly conclude that executive compensation is determined by their talents? Some of them (HP comes to mind) are frequently led by a bunch of no talent assclowns, where the company would probably be better off without any CEO at all, yet they'll frequently still receive higher salaries in a year than most people will make in a lifetime.

* And for some more anecdotal data: Paul Otellini (Intel) serves on Google's board. Eric Schmidt (Google) served on Apple's board. Intel's board of directors has 11 people, and includes one former CEO, two people who serve on three other boards, three other CEOs, and their own CEO. Walmart's board is similar: several former or retired CEOs, some still serving CEOs, and several people otherwise involved in the management of other companies. If you look at PepsiCo, you see it again: only one member who is not a current or former member of the executive leadership of a company. Similar trends exist for Microsoft, Bank of America, Yahoo, Verizon, ExxonMobil, DuPont, and.. Do I really need to keep going?

There were sources there backing up my argument. The graphs were the opposition to his point- the rest of it was support for my point.

EDIT: And I don't know what words I'm using wrongly, but I never mean to imply absoluteness to my arguments. I don't think CEO's have complete control over their salary, but that they have a significant amount of influence over it, enough so that you can't say that the "pricing system" has determined their value. If a fair pricing system did determine their value, then their pay would logically follow much fairly closely with performance. The data I have provided shows that it does not.

EDIT: For example: WHY do the charts focus so much on revenue? A company can make profits while its revenue drops: for example, if they downsize and lay off a fair number of workers, profits go up and the company can potentially become more sustainable despite a drop in revenue. Ultimately, the random nature of those graphs only tells me that a more complicated model than "pay vs revenue" needs to be considered. Clearly, we aren't getting the real picture from those graphs.

The Bloomberg graph covers returns (defined by them at the bottom as "Total return included both price appreciation and reinvested dividends of a company's stock in its 2009 fiscal year"), not revenue. The Indystar article dealt with shareholder returns. Only the Massachusetts data deals with revenue.

[Vaniver]

Capital owners, such as CEOs,

No. Just stop right there.

CEOs are managers. In many modern corporations, management and ownership are almost entirely distinct, which creates a whole host of issues collectively known as principal agent problems or agency cost.

CEO compensation is a cost for capital. Take a look at your Figure 9 from the previous page, and focus on 2000-2001. When you look to the left, what do you see? Massive CEO compensation and declining corporate profits. Look to the right, and what do you see? Mostly stagnant (though very volatile) CEO compensation and rising corporate profits.

The power that CEOs have over CEO compensation determination is a well-known and well-discussed problem. But unless there's a way to determine the value a CEO adds to a company, it's impossible to say if a CEO negotiation technique that raises their wage increases or decreases efficiency. Most of the work that I've seen that attempts to determine value, rather than just express outrage, finds CEO pay is reasonable or near reasonable.

For example, looking at your most recent graph, there's a clear positive correlation. It's not a very strong correlation, but that's probably because you're only looking at one factor.

[Ghostbear]

CEOs are managers. In many modern corporations, management and ownership are almost entirely distinct, which creates a whole host of issues collectively known as principal agent problems or agency cost.

By technical definition, yes. Would you challenge the idea that most CEO's see a lot of their compensation in the form of stock options (aka, capital)? That, in general, CEOs will be some of the larger capital owners of a corporation they work for? Being a CEO doesn't make you a capital owner solely by virtue of that fact, but I'd say you can guess with a high degree of certainty that a CEO is going to be a capital owner. The fact that they are often a large capital owner is part of why they have power over their compensation.

CEO compensation is a cost for capital. Take a look at your Figure 9 from the previous page, and focus on 2000-2001.

There are so many links, images, and sources on the previous page- "Figure 9" gives me zero information to determine what I should be looking at. This isn't a textbook.

The power that CEOs have over CEO compensation determination is a well-known and well-discussed problem. But unless there's a way to determine the value a CEO adds to a company, it's impossible to say if a CEO negotiation technique that raises their wage increases or decreases efficiency. Most of the work that I've seen that attempts to determine value, rather than just express outrage, finds CEO pay is reasonable or near reasonable.

And the work you've seen hasn't been provided as evidence for this discussion, while some of mine has been. "I've totally seen evidence that says you're wrong" isn't compelling, you're not Fermat.

As well, CEOs having a large amount of power over their own pay would indicate to me that their pay is only reasonable when taken in the context of "compared to other CEOs" or "in the CEO's opinion"- which isn't how you determine if it is reasonable. If workers had half as much power over their pay, it'd be high unreasonable high too. Why are CEO's exempt from this? Would you say that executive compensation was unreasonably low before the late 90s?

Spoiler:

For example, looking at your most recent graph, there's a clear positive correlation. It's not a very strong correlation, but that's probably because you're only looking at one factor.

Uh, I'm definitely not seeing a "clear" positive correlation. Are you looking at the same chart as I am? Even if there is, a "not very strong correlation" is proving the point: you can't predict CEO salary based on CEO performance with any measure of accuracy.

Anyway, your assertion that CEO pay is fair because it's determined by the pricing system seems completely absurd in the face of:

The power that CEOs have over CEO compensation determination is a well-known and well-discussed problem.

When they are a significant part of the pricing system themselves, I am not willing to accept "the pricing market has spoken, and thus is it good".

Double anyway: I have provided my argument and evidence for why I think:

But, the proportionate share is determined by the price system. You can't just say "too much" is going to capital- how would you determine the proper share for capital?

Is bullshit. If you aren't willing to defend that statement, then lets go back to the discussion of labor compensation vs. capital compensation.

[Vaniver]

Would you challenge the idea that most CEO's see a lot of their compensation in the form of stock options (aka, capital)?

I wouldn't challenge the fact, but I would challenge the implication. When talking about production, it doesn't matter if someone is compensation in cash or stock or gold or pickled fish, what matters is what they add to the production process. In many modern corporations, every employee receives some of their compensation in the form of stock or stock options. That doesn't mean there's only capital involved in their production.

CEOs are paid by the period for the time, not the money, that they invest, and so they're labor. The story of stagnant median real wages and growing top X% wages is not that capital wins and labor loses, because part of labor is getting significant gains. It's that people whose labor is multiplicative win and the people whose labor is additive stay the same, because as the world grows wealthier multiplication is worth more than addition.

There are so many links, images, and sources on the previous page- "Figure 9" gives me zero information to determine what I should be looking at. This isn't a textbook.

I apologize for presuming you would be familiar with your own posts. The image is here.

Why are CEO's exempt from this? Would you say that executive compensation was unreasonably low before the late 90s?

No. What changed between the 80s and the 2000s was that firm size shot up, and the compensation for executives is influenced in several ways by firm size. You can read Gabaix's paper about that here.

Uh, I'm definitely not seeing a "clear" positive correlation. Are you looking at the same chart as I am? Even if there is, a "not very strong correlation" is proving the point: you can't predict CEO salary based on CEO performance with any measure of accuracy.

I am looking at the same chart; the correlation is .094. (The chart excludes an outlier which makes the correlation negative, but since we're comparing our abilities to read charts, I excluded it from the analysis.) That implies there is information stored there, so you can make predictions with some positive accuracy.

But, most importantly, showing that two metrics are only weakly coupled doesn't mean that one of those metrics is flawed. When you look at the what they're actually comparing, it's surprising that the correlation is as high as it is. They compare shareholder return over 2009 (i.e. stock price at 2009 minus stock price at 2008 plus any dividends) to the difference between executive compensation over 2009 and over 2008- but if a company had a great year in 2008 and a good year in 2009, and the CEO is compensated entirely based on shareholder return, it would show a high shareholder return and a declining CEO compensation, contributing to a negative correlation even through CEO compensation is entirely determined by shareholder return.

Anyway, your assertion that CEO pay is fair because it's determined by the pricing system seems completely absurd in the face of:

Let me rephrase my earlier statement: it's a well-known and well-discussed problem for shareholders.

Price is a negotiated relationship that determines how much of the surplus value from trade goes to which party. So long as both parties get out more than they put in, the trade is a fair one. For trades that should happen, that means there are a range of prices that should be acceptable to both parties, and negotiation determines who gets what share.

Low-friction markets give both parties an opportunity cost- if I am willing to pay $10 for a banana, and you sell them for $5 while your friend sells them for $1, even though I'd be willing to trade with you in the absence of your friend, because your friend is offering a better deal trading with you represents a loss. This constrains the range of acceptable prices, and in perfect markets in equilibrium that price has a range of 0- i.e. it's a single value.

The executive compensation committees and board-stuffing trickses are good for CEOs and bad for shareholders. It's not clear that, overall, they make the gains for bidding for a great CEO negative for shareholders. When they lower friction in the CEO market, they're a good thing for overall efficiency, but when they represent coercion and perverse incentives, they're a bad thing for overall efficiency. I also suspect it's better to encourage large, long-view shareholders to take a more active interest in the CEO selection and compensation process than to restrict the CEO's bargaining tools.

(One of the things to note about CEO compensation- it depends not just on the company that hired them, but the company that might hire them away. That's one of the reasons why Japanese CEOs are compensated so poorly compared to English-speaking CEOs; the labor market is much higher friction and smaller. English and Australian CEOs are compensated more highly than similar CEOs in other countries because they come from the Anglosphere, and thus could (and often are) hired by American companies.)

[Ghostbear]

I wouldn't challenge the fact, but I would challenge the implication. When talking about production, it doesn't matter if someone is compensation in cash or stock or gold or pickled fish, what matters is what they add to the production process. In many modern corporations, every employee receives some of their compensation in the form of stock or stock options. That doesn't mean there's only capital involved in their production.

CEOs are paid by the period for the time, not the money, that they invest, and so they're labor. The story of stagnant median real wages and growing top X% wages is not that capital wins and labor loses, because part of labor is getting significant gains. It's that people whose labor is multiplicative win and the people whose labor is additive stay the same, because as the world grows wealthier multiplication is worth more than addition.

You're arguing the purpose of a CEO- I was saying that CEOs often represent significant capital ownership, due in part to their compensation method. You don't appear to disagree with that, and I'd like to attempt to surface this discussion back to its roots of labor vs. capital returns with respect to productivity gains, so I'm not going to delve in discussing any of the other points raised.

I apologize for presuming you would be familiar with your own posts. The image is here.

I originally had an unnecessarily snarky way of saying this, then I realized it wasn't going to help the situation, so I'm going to ask kindly yet bluntly: Please don't be condescending to me. Being familiar with my own posts does not involve remembering a completely incidental numerical label applied to an image by somebody else. Implying that I'm not familiar with my posts because I don't remember what is meant by "Figure 9" is ridiculous and insulting. Our discussion has not been warm, but if we're going to make it rude, then I have no desire to participate. If I have had any implied rudeness earlier in our discussion then I apologize (though it was not intentional- not that that's a sufficient excuse anyway), and again request that we keep condescending and rude exchanges out of it.

As for the 2000-2001 frame: what of it? CEO pay dipped a bit from being a ~400% gain over 10 years to being a ~150% gain over 10 years (compared to a ~4% gain for a production worker). The dip is also nearly perfectly tracked with the S&P 500, which probably represents a significant share of CEO compensation (since, as you yourself said, executive compensation rises with corporation size). I don't see it invalidating anything.

I am looking at the same chart; the correlation is .094. (The chart excludes an outlier which makes the correlation negative, but since we're comparing our abilities to read charts, I excluded it from the analysis.) That implies there is information stored there, so you can make predictions with some positive accuracy.

But, most importantly, showing that two metrics are only weakly coupled doesn't mean that one of those metrics is flawed. When you look at the what they're actually comparing, it's surprising that the correlation is as high as it is. They compare shareholder return over 2009 (i.e. stock price at 2009 minus stock price at 2008 plus any dividends) to the difference between executive compensation over 2009 and over 2008- but if a company had a great year in 2008 and a good year in 2009, and the CEO is compensated entirely based on shareholder return, it would show a high shareholder return and a declining CEO compensation, contributing to a negative correlation even through CEO compensation is entirely determined by shareholder return.

A correlation of 0.094 is incredibly weak, and indicates practically nothing. It certainly isn't a "clear, positive correlation", and it definitely isn't enough to be clear when visualized. If you expect CEO pay to follow CEO performance, I'd say there should be a correlation that is better than pitiful. I'd also expect your hypothetical scenario to not have been too common through the S&P 500 for this chart's time frame: it went from 903.25 on Dec. 31, 2008, to 1,115.10 on Dec. 31, 2009- a 23.5% gain. Sure, it'd still happen to some, but if there was a strong relationship, you would expect those cases to be averaged out over such a large data set.

Price is a negotiated relationship that determines how much of the surplus value from trade goes to which party. So long as both parties get out more than they put in, the trade is a fair one. For trades that should happen, that means there are a range of prices that should be acceptable to both parties, and negotiation determines who gets what share.

No, it's not "fair" per the common usage of the term ("marked by impartiality and honesty : free from self-interest, prejudice, or favoritism"). It's a mutually beneficially trade, but not a fair one- fair doesn't make any sense in this context. To return to the point I was trying to make (since I feel this giant descent into nitpicking is pointless for everyone present), just because labor is benefiting from their wages does not mean that they are getting a fair share of the gains from productivity that none of the parties involved were party to the creation thereof.

Say that an employee values their time at $20 / hour, and a company values that employees time at $1,000,000 / hour, in this case, if the employee was paid $25 / hour, both parties would be benefiting from the situation (per hour, the individual has effectively gained $5, and the corporation has gained $999,975), but it would not generally be a fair transaction: the vast majority of gain is only going to one party. The situation, if reversed- the employee is paid $999,975 / hour- giving them a gain of $999,955 and the company a gain of $25- is not fair either. For it to be fair, it'd need to meet somewhere closer to the middle (where exactly would depend on several factors, such as how difficult it'd be to replace that employee). Such is the case here: labor and capital have gotten a mutually beneficial gain from that increased productivity, but the lion's share of that gain has been concentrated with one group. And to reiterate the point that originally started this whole sidetracking: saying that that's what the pricing market has determined to be right is not an ample argument against that pricing market being broken.

* In fact, I'm not sure any of the definitions of fair beyond "conforming to the established rules" could possibly fit in this scenario. But when the argument is that the established rules are not, err, fair, it's not particularly helpful for the discussion.

[Vaniver]

it definitely isn't enough to be clear when visualized.

I have looked at a lot of scatterplots in my life, and I was able to tell the correlation was positive from looking at it. (It's clearer on the large screencap than cropped version.) I was confident enough that I took the time to copy the data into a spreadsheet and run the correlation.

If you expect CEO pay to follow CEO performance, I'd say there should be a correlation that is better than pitiful.

What is CEO performance? What if shareholders decide that the CEO should be paid based on customer satisfaction, which predicts long-term growth but not short-term stock price? What if they decide that the CEO should be paid based on 5-year return, instead of 1-year return? What if they decide the CEO's pay should only depend on size of the corporation, and not on stock performance, so that they can attract appropriate talent but not give them perverse incentives to game stock metrics?*

In all of those cases, I would expect the correlation between the difference of 2 years of pay and the difference of stock prices one year apart to be very low. The data we're working off of is a poorly designed empirical test of whether CEO pay and performance are coupled.

*This is probably a poor way to do that, since the CEO will probably be far too eager to acquire other companies.

it's not "fair" per the common usage of the term

Why would you expect a definition that includes "free from self-interest" to apply to prices?

where exactly would depend on several factors, such as how difficult it'd be to replace that employee

Right. How does this differ from the price system, again? What we're seeing is what you would expect if the median laborer were easy to replace, the upper decile of laborers were difficult to replace, executive-level talent ever more difficult to replace, and capital also hard to replace. Does that seem plausible?

[Ghostbear]

What is CEO performance? What if shareholders decide that the CEO should be paid based on customer satisfaction, which predicts long-term growth but not short-term stock price? What if they decide that the CEO should be paid based on 5-year return, instead of 1-year return? What if they decide the CEO's pay should only depend on size of the corporation, and not on stock performance, so that they can attract appropriate talent but not give them perverse incentives to game stock metrics?*

In all of those cases, I would expect the correlation between the difference of 2 years of pay and the difference of stock prices one year apart to be very low. The data we're working off of is a poorly designed empirical test of whether CEO pay and performance are coupled.

*This is probably a poor way to do that, since the CEO will probably be far too eager to acquire other companies.

I'd say that those are fairly unrealistic shareholders compared to those of the modern day- the focus on short term profit in the modern era is oft criticized. CEO performance is the ability to increase the value of company for the shareholders.

Why would you expect a definition that includes "free from self-interest" to apply to prices?

Because there are two parties involved, with one having a far more significant influence over the process than the other. I would definitely consider "free from self-interest" essential to the outcome being fair- otherwise the party with more power (often for reasons unrelated to the actual value) will give themselves far more of the gains than the other.

Right. How does this differ from the price system, again? What we're seeing is what you would expect if the median laborer were easy to replace, the upper decile of laborers were difficult to replace, executive-level talent ever more difficult to replace, and capital also hard to replace. Does that seem plausible?

The price system is including factors other than just that. It will include factors completely separate from the value of each party: a company can more readily afford to not gain their expected gains from the employee (even if the maximized scenario, e.g. paying them nothing) than the employee can (even in the minimalized case of being paid nothing, because that'd allow them to add work experience to their resume). That has nothing to do with the value of either party, but it affects the bargaining power in a very real way. In short, the pricing system is broken, as I've been saying. One party has successfully acquired far more bargaining power than the other due to outside factors, and is not using it to grant themselves nearly all of the gains, and nearly none of it to labor. Or would you say that capital was just giving labor more of those productivity gains pre 1980 out of the kindness of their hearts?

I would consider quite a few degrees to fall under "median" labor, yet they can be quite difficult to replace- I often see it in the news that companies are annoyed with the lack of workers with specific skill sets that they need. I don't consider executive level talent hugely difficult to replace either.

[Thaciv Yovero]

"People criticize shareholders for being short term thinkers, therefore the fact that CEO pay has an obvious but not extremely strong correlation with a short term indicator of shareholder value shows that CEO's are not being compensated properly."

How is that a valid argument?

[Ghostbear]

"People criticize shareholders for being short term thinkers, therefore the fact that CEO pay has an obvious but not extremely strong correlation with a short term indicator of shareholder value shows that CEO's are not being compensated properly."

How is that a valid argument?

Your criticism would be more pertinent if it was of the argument I was making.

[Thaciv Yovero]

That's the exact argument you are making.

[Ghostbear]

Well, I'm certain you know my own arguments far better than myself. Congratulations, you win.

[Thaciv Yovero]

Well, I'm certain you know my own arguments far better than myself. Congratulations, you win.

Perhaps instead of getting snarky, you could state how I've misrepresented your argument. I'm not sure why you'd think I'd just say "Oh well, if he says it's not his argument, then I guess I should just completely disregard his earlier posts."

You've said in this thread:

1) "The CEO's of the S&P 500 ("biggest corporations) appear to have no substantial link to pay increases and company performance. "

In other words 'There is not a strong correlation between CEO pay and this short term measure of performance, therefore CEO's are not being compensated properly.'

2) " the focus on short term profit in the modern era is oft criticized. CEO performance is the ability to increase the value of company for the shareholders."

In other words 'People criticize shareholders for focusing on short term results.'

Put the two together and you have 'People criticize shareholders for focusing on short term results, therefore the fact that CEO pay is not strongly correlated with a short term indicator of shareholder value shows that CEO's are not being compensated properly.'

You've used a clearly flawed evidence to back up your argument, and then you responded to criticism of the flawed evidence by saying, basically 'People criticize shareholders for focusing on short term results'. I don't see how I misrepresented your argument at all.

[Vaniver]

a company can more readily afford to not gain their expected gains from the employee (even if the maximized scenario, e.g. paying them nothing) than the employee can (even in the minimalized case of being paid nothing, because that'd allow them to add work experience to their resume). That has nothing to do with the value of either party, but it affects the bargaining power in a very real way.

Why doesn't it have anything to do with the value of either party? Stability is a benefit that companies bring to the table which employees rarely do, and when the employees do, they get a larger share because of it.

Or would you say that capital was just giving labor more of those productivity gains pre 1980 out of the kindness of their hearts?

No, I think labor had more value then, and I think that in a global context America before 1980 was a golden period for labor that will not come again anytime soon, and shouldn't be expected to.

I don't consider executive level talent hugely difficult to replace either.

Then I recommend going into headhunting- you can turn your skill in finding executive level talent into significant amounts of cash whenever you match up a candidate and a job.

[Ghostbear]

Perhaps instead of getting snarky, you could state how I've misrepresented your argument. I'm not sure why you'd think I'd just say "Oh well, if he says it's not his argument, then I guess I should just completely disregard his earlier posts."

[...]

You've used a clearly flawed evidence to back up your argument, and then you responded to criticism of the flawed evidence by saying, basically 'People criticize shareholders for focusing on short term results'. I don't see how I misrepresented your argument at all.

As I said in my PM, I'm not interested in definition wars or circular quote dances- if I'm going to respond to criticism of "this is what your argument is" I need to have some idea of how you reached that conclusion. This gives me something to work with. Your interpretation of my argument is flawed because you're taking two disjointed statements and saying that they directly connected. Shareholders focusing on short term results was directed at potential alternative definitions for CEO performance. You're also hugely misrepresenting my argument with respect to executive compensation: they aren't being compensated properly because they have a huge influence over their compensation. Performance not having more than an infinitesimal correlation with compensation was a response to others saying that they're being compensated properly because their pay is related to their ability to enrich the shareholders.

Why doesn't it have anything to do with the value of either party? Stability is a benefit that companies bring to the table which employees rarely do, and when the employees do, they get a larger share because of it.

I'm not certain how "I can't afford to not buy food or pay my rent" vs. "By nature of our large size, we can afford to lose a large amount of productivity" is stability. It's a power dynamic- the fact that the corporation can afford larger and longer lasting gaps in acquiring those gains is because the company is larger. Should we reward an entity solely for being larger, or should we reward it for the positive traits that allowed it to do become so? I don't see any logical reason to give a company a larger share of the gains of an employees work just because the company is worth more than the employee.

No, I think labor had more value then, and I think that in a global context America before 1980 was a golden period for labor that will not come again anytime soon, and shouldn't be expected to.

Because we shouldn't expect it to come back means we should just give up on it? "Hey, things were great for you before, but not anymore, sucks to be you (unless you're already rich)" ? Every problem becomes acceptable if we just agree to lie down and do nothing about it.

Then I recommend going into headhunting- you can turn your skill in finding executive level talent into significant amounts of cash whenever you match up a candidate and a job.

This is a completely useless rebuttal. "Not hugely difficult" is not the same as "I can do it and am talented at such". Drawing stick figures is not hugely difficult- that doesn't mean I'm any good at it.

[Vaniver]

I'm not certain how "I can't afford to not buy food or pay my rent" vs. "By nature of our large size, we can afford to lose a large amount of productivity" is stability.

Okay. Take five minutes, think about it, and see if you can come up with some reasons why an entity would prefer to trade with someone resistant to shocks than someone susceptible to shocks.

Because we shouldn't expect it to come back means we should just give up on it?

...Yes? Why is this a question? When a battery runs out of stored energy, it's silly to hope that it'll spontaneously recharge.

This is a completely useless rebuttal. "Not hugely difficult" is not the same as "I can do it and am talented at such". Drawing stick figures is not hugely difficult- that doesn't mean I'm any good at it.

If you were good at drawing stick figures, then I would be interested in your opinion of its difficulty, especially if you were good at drawing stick figures and making paintings, because then you could compare the two. When you're not good at drawing stick figures, then it's likely that there's some hidden complexity that makes it challenging that you're not aware of until you become good at it, or at least try to make a few.

For example, one could go into an art museum, look at a Pollock piece on the wall, and say "That's not difficult!". The analogy to my rebuttal would be "then you should make some yourself; Pollock is making quite a bit of money!" Were you to take that advice and attempt to actually do it, you would quickly discover that there are skills Pollock has which you do not, and thus it is difficult to get your work accepted into a museum.

Going a level deeper, consider this as the difference between rationalism and empiricism. Your objections are typically along the lines of "I don't see why X" or "I don't understand X," which are fine as statements of ignorance but troublesome as objections. You don't consider this a hard problem? What would it mean for it to be a hard problem? Is that what we see?

The reason I like the price system so much is because it provides feedback. If headhunting is an easy problem, but people are being paid a lot to do it, that means there are too few headhunters. But even better, the solution is built into the problem: people who can hunt heads will switch jobs because headhunting pays better- dropping the price of headhunting.

[Zamfir]

Ghostbear, Vaniver, you might be interested in this article: http://hbr.org/2012/03/the-incentive-bubble/ar/1?referral=00134

It's an argument, in the harvard business review, against current payment structures in the American business world. The writer does not appear to have a principled objection to paying managers boatloads of money. His claim is that the incentive structures for both top-level managers and financial fund managers were intended to replicate the incentive structure of company founders, but fail to do so for a variety of reasons. Even you disagree on the 'failure' part, I think he makes a good point that increased compensation for company managers and in the financial sector are two sides of the same phenomenon.

Somewhere int he middle he summarizes his points as follows:

In short, we have come to evaluate and compensate managers on both sides of the capital market as if the market could precisely disentangle skill from luck. Professional sports provide a common and convenient metaphor for business and financial managers. But distinguishing skill from luck is relatively simple in sports. The success of Roger Federer or LeBron James comes almost exclusively from aptitude, hard work, skill, and expertise, with only an infinitesimal amount of luck involved. We pretend that we can assess managers and investors with the same precision through financial markets, when in reality that ideal level of measurement is unobtainable and current compensation arrangements don’t even try to approximate it very seriously.

We might not worry about all of this if market-based compensation resulted merely in payments to individuals who are skilled at marketing themselves while not actually adding any financial or social value. Indeed, we reward such individuals all the time in product markets. But the fact that both sides of the capital market have become captive to the financial-incentive bubble is highly problematic for three reasons.

First, the inefficient risk taking engendered by these incentive contracts has widespread consequences for the allocation of capital in our society. Indeed, as detailed above, the recent financial crisis is the latest and largest manifestation of the disruptions a financial-incentive bubble can create. Significant spillovers arise when individuals are given such asymmetric incentives to pursue risk. Second, talent will continue to be misallocated in our economy as long as outsize rewards are available in certain professions. Third, the surge in income inequality that troubles many people today can be traced to windfalls for managers and investors and the rise of alternative assets. As a result of these generous contracts, the top 0.1% of the income spectrum is dominated by executives and financial professionals.

[Vaniver]

Zamfir: Thanks for the link! I agree that there's much to dislike about modern corporate American governance. (It's worth mentioning that it outperforms other measures of governance by a wide margin, including most of the things people suggest as replacements, but there are many improvements that could be made.)

[nitePhyyre]

It's worth mentioning that it outperforms other measures of governance by a wide margin, including most of the things people suggest as replacements

[Citation Needed]

[Vaniver]

[Citation Needed]

This seems like a good place to start.

[Princess Marzipan]

There should a forum rule against responding to [citation needed] with paywall'd papers.

[TheGrammarBolshevik]

Maybe it's just a university Internet thing, but I see no paywall.

[Princess Marzipan]

Yeah that would be your HARVARD Internet allowing you to bypass the paywall.

[addams]

Have any of you read or listened to what Elizabeth Warren had to say about where the money went?

She and a team of very bright people looked at the books. When asked point blank, "Where did the money go?" She responded with, "I don't know."

http://en.wikipedia.org/wiki/Elizabeth_Warren

The US went from being in the black to being in the red. That happened very quickly. It was not money spent on the welfare of the people. Just the opposite.

Everyone wants to understand the world that we live in.

If, Elizabeth Warren can't figure it out, then, what chance do we have?

Yes. The conservatives in the US are extremely hypocritical. In my experience these people tend toward magical thinking and also look for an easy way to blame the ills of the world on someone that is unable to defend themselves.

ech. The poor are becoming more poor and the rich are becoming more rich. It sucks.

The argument can be made that Jesus would bless a nation that cares for its people. That is the only way that I can think of to reach inside the bubble of fear and blame and greed. Numbers are not going to do it.

Maybe a fear of God would. But; Not likely.
The battle lines have been drawn. Either you are with them or you are the enemy. There seems to be no compassion for the enemy within the conservative camps.

The presumption of innocence is gone. It was always a difficult concept. Lazy minds don't want to be bothered with it.
Where did the money go? It did not go into benefits for the people. The amount that disappeared can not be accounted for nor replaced with regular arithmetic.

The power of self determination is to some degree regulated by money. Freedom from want was one of our common ideals. We lost that in this war, too?

Fuck.