In a recent blog post, I wrote:
Aside for non-economists: Why would reductions in income tax rates on corporations and on high-income individuals even be expected, at a theoretical level, to increase real wages? By increasing the incentive to invest in capital. The greater the capital to labor ratio, the higher are real wages.
Commenter Robert asked:
Could someone expand on this a bit for me? Does investing in capital, mean investing in capital goods (i.e., land, machinery, tools, etc)? I don’t understand why that would necessarily lead to higher real wages.
Commenter Laron gave a nice succinct answer:
The capital goods like machinery increase labor’s productivity, which increases wages. Others can chime in with more detail or to correct me tho.
Here’s what the article on “Capital Gains Taxes” in The Concise Encyclopedia of Economics says about the issue:
Between 1900 and 2000, real wages in the United States quintupled from around fifteen cents an hour (worth three dollars in 2000 dollars) to more than fifteen dollars an hour. In other words, a worker in 2000 earned as much, adjusted for inflation, in twelve minutes as a worker in 1900 earned in an hour. That surge in the living standard of the American worker is explained, in part, by the increase in capital over that period. The main reason U.S. farmers and manufacturing workers are more productive, and their real wages higher, than those of most other industrial nations is that America has one of the highest ratios of capital to worker in the world. Even Americans working in the service sector are highly paid relative to workers in other nations as a result of the capital they work with. In their textbook, Nobel laureate Paul Samuelson and William D. Nordhaus noted: “Because each worker has more capital to work with, his or her marginal product rises. Therefore, the competitive real wage rises as workers become worth more to capitalists and meet with spirited bidding up of their market wage rates.” The capital-to-labor ratio explains roughly 95 percent of the fluctuation in wages over the past forty years. When the ratio rises, wages rise; when the ratio stays constant, wages stagnate.
The quintupling in the above is certainly an underestimate because the Consumer Price Index, used to adjust for inflation, overstates inflation.
A way to think about it is with a person on a desert island catching fish. If he does it with his bare hands, he can catch, say 2 fish a day, enough to keep from starving. But if he fashions a stick that can spear the fish, he can catch, say 4 fish a day. So that one piece of capital, the spear, has doubled his productivity. His real wage has doubled.
Then Robert followed up:
Thanks Laron. That was kind of what I was assuming was meant in the quote. I don’t know if I fully trust a business to return higher productivity back to workers in the form of higher salaries, but I can see how and why that could happen.
Here’s why it would happen. A worker becomes more productive, not just with his current employer but also with other potential employers. So if an employer does not pay the employee an increased wage for increased productivity, another employer will offer more than the current employer. That will happen until the marginal revenue product (the marginal revenue produced by the employee’s marginal product) equals the wage.
READER COMMENTS
AJ
Oct 4 2020 at 10:39pm
This seems like too strong of a statement to make, especially since your argument why it will happen has assumptions baked into it. Will it happen? If all of the antecedent conditions are met, it’s certainly possible, but I can think of scenarios where it may not be the case.
Jon Murphy
Oct 5 2020 at 8:54am
There are certainly scenarios where it may not be the case the worker goes to another firm for higher wages. Knowledge and information are incomplete, after all. But remember: it is in both the worker’s and the competing employer’s interest to gain that necessary information:
If the competing employer sees his competition doing well (in technical terms, earning economic profit), he’ll want to figure out why and how he can improve his own firm to capture some of that profit.
The employee, likewise, has an incentive to constantly test his own value on the market and gain as much information and knowledge as possible.
With websites like Glassdoor and industries devoted to headhunting, the information gap is likely smaller now than it ever has been.
AJ
Oct 5 2020 at 11:27am
Jon –
I largely agree with your post. But I will say the incentives to which you are referring – to think in terms of those particular choices – are largely presupposed by economics textbooks. Those incentives and choices are framed by a person largely because of his or her education in economics and/or the kind of culture in which we live. A person with a different background, using a different form of rationality, or different understanding of human nature that contradicts that of economics textbooks may not frame his or her situation in such a way that results in making this type of choice.
Is the incentive always there? I don’t think so, but it certainly is if one is trained to think this way.
David Henderson
Oct 5 2020 at 11:35am
The desire to make money preceded by centuries the formal study of economics.
AJ
Oct 5 2020 at 12:18pm
OK
Jon Murphy
Oct 5 2020 at 11:59am
I think it is a broad and general rule of humanity that people, in general, wish to improve their situation. Are you aware of any culture where people systematically whish to deteriorate their current situation? It seems to me any such culture would not long last.
Yes, economists presuppose people want to improve their lives. That seems like a reasonable assumption to me supported by pretty much all of human history.
AJ
Oct 5 2020 at 12:17pm
To frame ‘wish to improve their situation’ by making choices the way David describes is one particular way of thinking about one’s life and is framed by one’s upbringing in economics and/or our capitalist culture. The impasse ‘should I assess my market value, apply to a new job and get an offer, or attempt to negotiate?’ is certainly a decision one can arrive at independent of the economic way of thinking, but it doesn’t always mean more money = good. One can certainly arrive, using a different set of assumptions than the economist and using a different form of rationality (not cost v. benefits), to the conclusion that leaving one’s job – even if he or she thinks he or she is ‘underpaid’ – for another higher paying job is bad, so there is no incentive to consider that path. The economic way of thinking would probably call this irrational – there are unexploited gains from trade – but those gains are also largely presupposed and people that use a different tradition or rationality would disagree.
Jon Murphy
Oct 5 2020 at 12:49pm
Not at all. Why would it be? EWOT explains that sort of behavior very easily:
-The transaction costs may be higher than the benefits from the switch (eg, you wouldn’t spend $1 to leave a job for a $0.01 raise).
-There may be non-monetary benefits to the job (eg, you enjoy the people you work with, better benefits package, etc).
In either case, the benefit to moving to a job with a higher wage is below the cost of moving.
Remember: costs and benefits are subjective. What looks like, to the outside observer, as a better opportunity may not necessarily be so to the person actually choosing (as an aside, this is why I dislike explanations that use “irrationality.” Reeks of “Man of System” thinking to me).
In short, the good economist would not look at the refusal to take a job with higher wages as irrational nor as a violation of theory at all.
Now, a key assumption of David’s comment is that all else is held equal. If there were two jobs identical in every single way except that one paid more than the other, we can predict with reasonable assurance that an individual will choose the higher-paying job over the lower-paying job. And empirically, we see this borne out across all of human history. So, either people do want to improve their lives, or they act as if they want to improve their lives. Either way, the assumption holds.
But you dislike the assumption that people want to improve their lives. What alternative do you propose? What do you suggest better explains people’s real and actual behavior?
AJ
Oct 5 2020 at 1:45pm
There’s too much to respond to in your post and too little time on my side, so I’ll focus on just one.
David’s way of thinking involves a very mechanistic view of humans by employing law-like claims about behavior by obscuring over people’s intentions, beliefs, goals, purposes, reasons for actions, etc. This, I think, is touching on the Austrian criticism of Neoclassical economics. Of course, my initial criticism of David’s post was that he said it “will” happen which I don’t think that’s the case. But regardless, that view is incompatible to an Aristotelian view of how human action is to be understood, for example, with intentions, goals, reasons for actions, etc as well as how acts ought to be done. In our modern emotivist culture, we say “no is from ought” and “facts are value free” or “ends are given”, but I think that is mistaken and at odds with older traditions of thought. What is good, bad, better, or worse were simply taken to be factual statements – either true or false – relative to human’s natural teleology (Nicomachean Ethics, for example), but in our modern emotivist culture today, that is not the case. And I don’t see good reason to abandon that framework for this modern way of thinking.
So it’s not that I dislike the assumption. But I think that framework is the right way to think about making decisions. That affects how one frames their choices (which doesn’t involve employing terms like ‘utility’, for example), how one reasons through an evaluation of the means to move towards that end (which involves the virtues), and whether there are good reasons for desiring, for example, to be rich (which I would argue there are no good reasons for desiring it). But our preferences to think in this way (as well as yours) comes from our background and culture. If you’re trained to think about your decisions by trying to equate marginal benefits with marginal costs, which I think is a modern western way of thinking, then your preferences will be solidified as such.
What do economists say about people that were raised to think about it in an Aristotelian framework or for those that see behavior as objectively good or bad or for those that were raised in poor non-Capitalist countries or countries with vastly different cultures than ours, such as in the east?
That’s more of a rhetorical question that I’d like to think through more.
But anyways, this will need to be my last comment so I can focus on other projects today. Thanks for talking through this.
Jon Murphy
Oct 6 2020 at 10:07am
Well, no. That is explicitly discussed in the model and in the Economic Way of Thinking. Intentions, goals, reasons for actions, what ought to be done are all baked into the model, even neoclassical models. Indeed, your initial objection explicitly rejects the intentions, goals, and reasons for action as “presupposed by economics.”
The intention, goal, and reason for action is to better one’s self. That is true whether we are discussing utility-maximization of Samuelson or of the purposive action toward a given ends of Mises. The Austrian objection is not that people do not try to improve their lives, but rather that central planning pre-supposes the knowledge necessary for such improvement even though it can only be revealed in the market process (see Hayek’s Use of Knowledge in Society or Lavoie’s National Economic Planning).
I must admit: I do not understand what your objection is at all.
Rob Rawlings
Oct 4 2020 at 11:49pm
It seems to be generally accepted (see https://www.epi.org/productivity-pay-gap) that wages stopped growing with productivity some time ago. Is this analysis flawed in some ways ? If not , how does this trend (since around 1980) fit with the “competition will bid wages up to the level of marginal productivity so all productivity gains will be reflected in increased real wages” theory?
Rob Rawlings
Oct 5 2020 at 12:10am
Looking at the chart in the link from my first comment more carefully I think it can probably be explained by the fact that the share of income going to ‘production/nonsupervisory workers’ has declined over the period in question which to be consistent with the the marginal p[productivity theory of wages would presumably mean that the productivity of workers not counted as ‘production/nonsupervisory workers’ has gone up much faster. Even assuming that the miscalculations in Consumer Price Index masks some of the increase in real wages I still find it intuitively unlikely that productivity between 2 groups of ‘workers’ diverged in the way it did in 1980 and am curious as to what may have happened.
Mark Z
Oct 5 2020 at 1:14am
This claim has certainly been been disputed, and I don’t think it’s widely accepted. Scott Winship had a critique the decoupling here: https://www.forbes.com/sites/scottwinship/2014/10/20/has-inequality-driven-a-wedge-between-productivity-and-compensation-growth/#2da125a2eb4e
His article is from a few years before the link you list above, but he found adjusting for factors like hours worked (the EPI used annual household income), and comparing productivity, comparing productivity and compensation in the same workers yielded very different results. Scott Sumner also wrote a post here about this last year.
robc
Oct 5 2020 at 9:22am
Russ Roberts discussing Simpson’s Paradox and how it applies to this:
https://www.youtube.com/watch?v=1DU2IT8rl6c
Jon Murphy
Oct 5 2020 at 8:56am
Mark cites a good article. The EPI’s charts are largely incomplete; they look only at monetary pay. Adding in different forms of compensation (health care, PTO, etc), there hasn’t been divergence at all.
Rob Rawlings
Oct 5 2020 at 9:58am
The Scott Winship article is indeed very good and comes up with some good reasons to question the EPI’s claim to have demonstrated a productivity/pay gap since 1980.
I am still left a bit curious however as to why the 6 reasons he gives to undermine the EPI’s interpretation of the data all combined together in 1980 to radically change the slope of the curve.
Bill
Oct 5 2020 at 11:12am
Marty Feldstein also looked at the relationship between worker productivity and compensation here:
http://www.nber.org/feldstein/WAGESandPRODUCTIVITY.meetings2008.pdf
Jon Murphy
Oct 5 2020 at 11:29am
The short answer to your question, Rob, is there has been a shift to more non-monetary compensation, both by legislation and preference, since the 80s. It’s one of the ways firms compete in the global world
Rob Rawlings
Oct 5 2020 at 12:49pm
Jon,
Is there data to support the theory that there has ‘been a shift to more non-monetary compensation’ to a sufficient degree to explain the EPI chart ? My intuition would be that it tends to be driven more by growing income inequality (between ‘production/nonsupervisory workers’ and the rest). This could reflect either greater relative productivity growth by non-production/nonsupervisory’ group, , or that political power has been used to divert a greater slice of the pie to the latter group (or a bit of both).
Jon Murphy
Oct 5 2020 at 1:16pm
Yes, there is some empirical evidence but I do not have it on-hand at the moment. I am afraid I am away from my library.
Eric Rall
Oct 5 2020 at 1:17pm
That chart uses inflation-adjusted numbers, and it uses different inflation adjustments for productivity (the NFB deflator) and for compensation (CPI-U). If you use the same inflation index for both lines, or if you just use nominal dollars for both, then there’s still a gap but a much less dramatic one.
Kevin
Oct 10 2020 at 5:49am
Rob, wages may not be increasing in step with productivity, but compensation is.
https://www.americanactionforum.org/research/does-compensation-lag-behind-productivity/
Thomas Hutcheson
Oct 5 2020 at 8:14am
Ultimately the argument depends on the tax cut either increasing savings and investment or making investment more productive. Reducing/eliminating taseson business would certainly increase capital productivity by reducing all the distortions introduced by business tax codes, But should the reduction in business taxation be allowed to reduce total collections? Should the reduction not be made up with taxes on personal income (or even better, personal consumption)? This seems to me to have been the failure of the “Tax Cuts for the Rich and Deficits Act of 2017.” It seems very unlikely to me that the reduction in taxes on high income (and increase in negative savings by government) would lead them not only to save and invest the total amount of the higher income, but to reduce their personal consumption as well.
Thomas Hutcheson
Oct 5 2020 at 8:36am
A different point.
“A worker becomes more productive, not just with his current employer but also with other potential employers. So if an employer does not pay the employee an increased wage for increased productivity, another employer will offer more than the current employer.”
This is not quite right. Productivity is average output per worker; wages are (supposedly) set according to marginal product. It would be nice on distributional grounds if the technology embodied in new investment increased the marginal product of workers, especially low-income workers, there is no necessary reason to expect technology to behave in that way. Maybe in did in 19xx to 19yy (yeh!) but not in 19yy to 19zz (boo!).
Alan Goldhammer
Oct 5 2020 at 10:06am
Surprisingly, I agree with what David has written. However, I wonder if anyone has tried to do a study of the impact of financial engineering on the loss of American manufacturing since 1980. I chose this year, not because of the election of President Reagan but rather the influence of Michael Milken and the rise of junk bond financing. It was about this time we saw lots of corporate takeovers, the establishment of private equity and hedge funds to levels heretofore unseen.
Companies were acquired using debt and assets stripped away. I presume that this led to a loss of capital investment and decreased productivity. If this research has not been done, it would make a good project for an enterprising economics student.
Phil H
Oct 5 2020 at 10:56am
I think the answer to that is that financial engineering should make capital and assets flow more freely to the places where they will be used most efficiently, so in fact they stimulate more investment, not less. All those stripped assets ended up somewhere, after all, and presumably in the hands of someone who could make them more productive, and so was willing to pay for them.
Jon Murphy
Oct 5 2020 at 12:00pm
Why? Theory would predict otherwise. If the “cooperate raiders” were earning profits, then they were moving said assets from lower productivity uses to higher productivity uses.
Alan Goldhammer
Oct 5 2020 at 3:25pm
The simplistic answer is that the capital is not being deployed but is going into the private equity company. The company is saddled with debt and then an IPO is set to try to maximize profit for the private equity group. The company is not adequately recapitalized and eventually goes bankrupt How is this efficient. There are lots of business school cases here.
Jon Murphy
Oct 5 2020 at 5:35pm
Ok, but my objection stands. If the resources are reallocated and profits are made, then the resources are being moved from lower productivity uses to higher productivity uses.
Laron
Oct 5 2020 at 12:02pm
Hey that’s me! Mahalos, that’s pretty validating for someone like me that’s taken an autodidactic approach to econ study :).
sk
Oct 5 2020 at 2:45pm
For sure, except those laid off due to being replaced by capital.
Jon Murphy
Oct 5 2020 at 3:25pm
Even then. In the short run, there would be people laid off. But in the long run, these people get absorbed into new jobs that are created and pay. Ultimately, we do not see an inverse relationship between capital and labor.
David Henderson
Oct 5 2020 at 3:27pm
You’re welcome, Laron. Keep ’em coming.
JoeR
Oct 5 2020 at 6:01pm
This doesn’t seem right to me. The employee hasn’t increased in productivity. The capital increased his productivity. So another company wouldn’t want the employee per se. They’d want the employee + capital. But assuming they already have employees, they could get the same by adding the same capital there. The employee wasn’t special.
I think if all companies were homogenous, they would all increase capital spending to get the higher productivity, and assuming the capital was less expensive then the increased profits (or why would they do it?), they would make higher profit. It’s not clear that the employee would definitely get a cut of that, but there is a bigger pie to get split up so maybe.
But companies aren’t homogenous, and neither is capital. An engineering firm can increase productivity with more compute servers, but a construction firm can’t. So I would expect some companies to find ways to increase productivity via capital in a more profitable way, and capital will move towards those firms. And because they are increasing profitability, those firms would likely expand their businesses, which would mean hiring more people. And that would mean offering a higher salary to outsiders and employees to keep them. So heterogeneity seems like the primary reason why increased productivity via capital would go towards employee salaries.
Jon Murphy
Oct 6 2020 at 10:00am
That is one possibility. They could also lure the employee to work for them (thus depriving their competitor of a well-trained employee).
JoeR
Oct 6 2020 at 4:25pm
Are you assuming the increased capital in the first company made their employees better trained? If so, I can see the argument. But there are lots of increases to capital that don’t make the employee better trained while still making them more productive. Just a few that I’ve seen in engineering: Faster computers, more computers, additional monitors, faster software, more licenses for software, new software that provides additional functionality, more lab space, new or better lab equipment, better real estate (less crowded, more meeting rooms, better cubicles, better location to where people live), and more.
None of those trains the employee to be better. In fact, they might get trained to optimize their work based on those capabilities. In the context of the hiring company, they may be less capable then their current employees.
zeke5123
Oct 5 2020 at 6:52pm
Curious for your thoughts on this in conjunction with the non-compete and other ways (e.g., equity awards) that tether people to jobs. I believe this is popular in Silicon Valley.
Presumably, those agreements are priced into the salary when the agreements are signed. So perhaps no harm no foul. But I wonder if people, especially at lower levels, take these clauses seriously / fully grasp the consequence.
However, it does solve the firm’s problem of investing in an employee only to see the employee leave. Wonder if you start to de-couple capital spending from an increase in real wages?
Michael Pettengill
Oct 6 2020 at 10:18am
Why is more capital built in high tax blue states than in low tax red states?
Its true red states lure factories from blue states, but the capital builders mostly remain in the blue states. Or turn red states progressively blue and taxes go up to meet the needs of capital builders.
And this happens across international borders, Japanese and German automakers building factories in red states keep not merely the vehicle engineering that must be done tied to factories, and the manufacturing process and equipment at home.
Which low tax state has produced highly capital intensive industry, other than mining. And mining is capital destroying, not building. What’s build with the fruits of mining is capital concentrated in blue states and blue cities. Ie, the steel machines and steel building are not build in Appalachia where the met coal was mined.
Mark Z
Oct 6 2020 at 11:20pm
“Why is more capital built in high tax blue states than in low tax red states?”
Is it? I’m not asking rhetorically, is gross capital formation measured at the state level? A quick google search didn’t help find any answers.
“Its true red states lure factories from blue states, but the capital builders mostly remain in the blue states. Or turn red states progressively blue and taxes go up to meet the needs of capital builders.”
It seems like you’re describing Simpson’s paradox without realizing it. If high taxes were essential to capital formation, why would manufacturers tend to move from high tax to low tax states in the first place?
Anders
Oct 8 2020 at 8:02am
I am not an economist, but my job involves applied economic analysis to inform policy and institutional reforms in developing countries.
In that environment, this type of reasoning is often dismissed – for two main reason. The first, I submit, is purely political – we do not like the notion of lowering taxes for the ultra-wealthy (the US is the only country I know where there is something akin to a critical mass of popular support for this, coupled with, to us Europeans at times bizarre rhetoric deifying entrepreneurs and job creators).
The second and more important is the assumption of strong market competition and a flexible economy that systematically makes the most out of human and capital resources. This is an assumption that simply does not hold – at least not in the short term, and in most developing countries in part not even in the long term. As communism fell, most new independent constituent states witnessed decades of negative sectoral reallocation – from capital intensive manufacturing to low value-added services and small-scale agriculture – coupled with rampant concentration of wealth, as rapid liberalisation without functioning market institutions led to rent-seeking on what might well have been unprecedented scale and rapidity.
What do we advocate, then, in a country that does not have the elements that could make capitalism work for everyone? Or, for that matter, how can we make the US health care sector work as well as, say, consumer electronics – where even the poorest can now afford more processing power than the whole world had a few decades ago?
Similarly, what about the case of Sweden – a country that combined high VAT rates and high personal income taxes with a liberal labour market and very low corporate taxes? This lead to substantial overinvestment among companies and households (especially into housing) – a bubble that rapidly deflated in the early 90s leading to a third of the population defaulting on mortgages and overnight interest rates of over 500% to defend the peg against speculation.
Sweden has since found a compromise – retaining relatively high levels of distribution, but also investing heavily in infrastructure and retraining of laid-off workers. But this, of course, only works because of a strong tradition of consensus, transparency, and cross-party commitment to the notion that a vibrant private sector is a precondition for growth and the welfare system overall. Couple this with an overzealous media with hordes of journalists trawling through the books and sounding the alarm whenever a Director General spends more than 30 USD on an official dinner, and the key to the relative success is a system that self-corrects.
That, however, is the exception that proves the rule – the US, arguably, is competitive not because the federal government is, but because of high levels of decentralisation, competition among states, and constitutional constraints ardently defended by the most hyperactive judiciary in the world (I think… nothing else this extreme comes to mind).
I welcome your thoughts on this – my problem is: I find liberal economic thought and the Austrian school compelling and convincing, but whenever I try to put them into practice, I find them to be little more than a pipe dream.
Comments are closed.