Many people would argue that nurses provide more value to society than investment bankers*, yet the latter position is paid significantly more than the former. This fact is often lamented. My question is: which mechanisms have been proposed to actually solve this perceived problem and achieve wages that are aligned with social value production (however one may define that term)?

Here is what I believe is the underlying structural issue (please correct me if you think I'm mistaken here): wages are tied to productivity in terms of monetary value. Aspects like ecological damage done or social value production are (mostly) ignored. [I believe that in economics, things that prices don't account for but possibly should are called externalities.] Jobs like investment banking produce a lot of monetary value, while jobs like nursing produce a lot of social but not a lot of monetary value - thus the difference in wages between the two.

For ecological damage, CO2 taxation is a proposed solution that works essentially by artificially adding (negative) ecological value production to monetary value production, and thus achieving a price that incorporates both. I could imagine a similar solution for incorporating social aspects, i.e., society might subsidize either the wages of people working in socially desirable but unprofitable jobs, or the price of the services they offer. (Society would thus "internalize" the externality "social good done".) Has something like this been proposed?

I am interested in all kinds of proposed solutions that attempt to solve the perceived structural problem.

(This is a follow-up to my previous question in which I was convinced by the many great answers that money is probably the most efficient means of communicating social value.)

*This is an arbitrary but somewhat popular comparison - for the purposes of this question it's irrelevant whether you and I believe that this is actually true.

  • 14
    The fundamental problem here is deciding just who gets to impose their definition of "social value" on everyone else. E.g. from my point of view rap "artists", many professional athletes, or FTM celebrities in general actually have negative social value (since I must waste time paging past articles about them in my news feeds), yet they get paid lots of money.
    – jamesqf
    Sep 17, 2020 at 0:02
  • 3
    In theory investment bankers actually can produce a lot of social value by fulfilling their original role in the economy - moving scarce capital where it is needed for growth. For example imagine a loan given to fund a new green technology which reduces emissions 30%. The banker would have played a crucial role. Modern investment banking has devolved into a lot of speculation. Sep 17, 2020 at 5:40
  • 1
    @jamesqf I agree that that's a very important issue, but in my mind that's only half of the problem - the other half being how to actually achieve the wage alignment.
    – Eike P.
    Sep 17, 2020 at 12:51
  • 2
    There are various ways to solve the problem: either you enforce it via a dictatorship of some sort, or you assume that market conditions have done a pretty good job of incorporating most people's ideas of social value.
    – jamesqf
    Sep 17, 2020 at 17:44
  • 1
    Employment is much more voluntary than the question assumes: you know how much money a nurse makes, and how much a banker makes, so you are “free” to decide which career to pursue. Some people choose professions that are emotionally rewarding and prefer that emotional reward to a high salary. That’s not a perceived problem to them.
    – blud
    Sep 22, 2020 at 21:06

5 Answers 5


I just stumbled upon an article that asks essentially the same questions (although much better formalized and much more clearly formulated!) and also discusses the potential solution I thought of (subsidies for 'socially beneficial' jobs).

Firstly, they define 'socially useful' as having a large 'spillover':

Economic research increasingly indicates some professions have “spillovers,” meaning that the social value of an individual’s work can be much higher, or much lower, than that individual’s compensation. The job market does not account for all social value.

Some spillovers are quite large. Given how much good teachers raise the eventual incomes of their students, we calculate that spillovers from teachers are twice as large as the salaries teachers are paid. The benefits from medical research are even larger, amounting to over one-fifth of total income in the U.S. On the other hand, some sectors involve “zero sum” endeavors, in which profits come at the expense of other market participants. Examples include excessive litigation or financial traders trying to beat the market.

Spillovers jam the signals in the economy, misdirecting talented students. This can lead to severe misallocations of the talent pool — arguably a nation’s most valuable asset. Recent work in macroeconomics indicates that when economies fail to allocate assets to their best uses, growth slows and incomes stagnate. Although economists have long focused on policies that promote the acquisition of human capital through education, the efficient allocation of the resulting talent is just as important.

They proceed to consider the option of taxing or subsidizing individual professions differently, based on their spillover. They essentially conclude that while in theory this would be a good solution and might have a lot of benefits both for society and the economy, it will be hard to implement in practice: how to draw boundaries between different professions? This will probably open up a whole new field of lobbying and loophole-seeking. They argue that the best solution is to do this on a case-by-case, which essentially means: wages should be significantly increased for teachers, researchers, nurses, etc.

Article in Harvard Business Review that I quoted from above: https://hbr.org/2017/10/what-if-socially-useful-jobs-were-taxed-less-than-other-jobs

Scientific article by the same authors on the same subject: https://www.journals.uchicago.edu/doi/10.1086/693393

Brief summary of numbers + research on externalities / spillover in different jobs: https://80000hours.org/2017/06/which-jobs-do-economists-say-create-the-largest-spillover-benefits-for-society/


As with everything, there are multiple factors at play here. In this particular case, the full explanation easily runs to thousands of pages. The short answer is that, in a free market, wages are aligned with social value production, and it only seems otherwise because any given individual is utterly incapable of cramming sufficient data into their personal three pounds of tapioca pudding to actually get more than a dim and fuzzy view of a small piece of it.

So the only way to analyze it is to break it into pieces and sort out the principles on which it functions. The simple, underlying math that, like a fractal pattern, results in amazingly complex shapes.

Let's take a look at your nurse and investment broker example to start with. It makes a great illustration about where the seeming dichotomy between "social" and "economic" value comes from.

You want to make sense of this, so you think about how important the services of a nurse would be to you. Gee, that seems awfully important. Like, you could actually die without it. By comparison the investment broker is a mere convenience. Something that makes a small part of your life easier, but you could definitely get by without it. "That doesn't seem right..." You think in dismay, "How could the market fail like this?"

But you're looking through a keyhole. The nurse has a job that's huuugely important to the people served... But how many is that? A quick search suggests that a reasonable workload for a nurse is between 25 and 35 patients.

The investment broker, meanwhile, is much less important to you in particular. But a single broker can easily have hundreds of clients. A decent-size firm can easily have thousands. So while the benefit per customer is lower, you're looking at a minimum of between four and forty times as many customers. In the modern era of automated trading it can easily be far more than that.

And it scales up. Sports players and rappers can make more than even investment brokers because, while they get even less per customer served, they have customers in the millions. Zuckerberg makes oodles of money not because any particular Facebook user is particularly profitable, but because there are over two billion of them.

So if you look at the aggregate. -- Not, "how much good can this one person do for one person," but "how much good can this one person do for everyone they work for." Not "What would the societal damage be if we got rid of all of them," but "what would the societal damage be if we got rid of one of them." Then the numbers start to make a lot more sense. The societal damage from getting rid of all nurses would be staggering. But there are over 4 million RNs, and only 1.2 million registered brokers of any kind (insurance, investment, and all the little sub-categories that fill out the paperwork.) (Incidentally, thanks for picking two professions that have registries, it makes research easier.) So while nursing as a whole is more important to society, the money we're willing to throw at it is also split more ways, and the average individual nurse has less of an impact on society than the average individual broker.

In a free market, the amount earned is directly related to societal benefit. People will generally only pay you to make them better off. Or, at least, less badly off than they would have been. They get cranky when you both make them worse off and demand they pay for the privilege. (This is not to say that monetary pay is the only measure of social benefit, just that being paid nearly always corresponds to having provided some perceived benefit to someone.) Furthermore the amount they are willing to pay you is directly related to how much they value your particular services compared to the other things they could have purchased with that amount instead.

All this conflicts with your social instincts because they were built for a tribal group of, at most, a couple hundred. The idea of providing a miniscule benefit to thousands of people... Well, in computer science parlance, it falls victim to an overflow error on the number of people and bails out of the multiplication process far too soon. It just doesn't compute correctly with numbers of people that you can't easily visualize.

The same thing is true of the so-called "externalities." The "things the market should account for, but doesn't." Sorry, but the market does account for those things, and has simply ranked them as less important than other things. Does that mean the market is wrong? Maybe. Markets aren't perfect. They can generate incorrect results, same as any other process. They simply generate the correct answer more often than any other method we know. So while you may occasionally be correct at identifying a market failure, the odds are not in your favor. (as evidenced by the percentage of business startups which fail.)

To apply this to the ecological example, people say they want to cut CO2 emissions because that's the currently fashionable way to talk... Then they go out and buy an SUV... Well, talk is cheap. If you go by what people say, then the current climate hot-button issues are one of the most important things ever. But if you go by what they actually voluntarily spend their money on? It seems like the climate activists have failed to truly convince most people.

Now, you can try to deal with this by collecting up what people say (we call it "voting") and then using force to compel them to prioritize what they said they thought was important despite it not being what's actually important to them once all the costs are tallied... I'm sure you can see where this runs into trouble...

Most of these "externalities" are things that people accept as important in the theoretical sense, but when they have to rank the importance against other concerns it quickly drops low. When you merely ask them things like "Would you like to lower CO2 emissions" they don't have any sense of what they'll have to give up in order to do it. Perceived benefit measured against a cost of zero! You'd have to be an idiot not to take that deal!

But if you start bringing actual cost into it, their answer changes. Because now they have something to compare against. Are they willing to give up the freedom of movement their personal vehicle provides? Are they willing to cram themselves into a tiny apartment? Stop heating and cooling their house beyond the bare minimum needed for survival and just deal with the discomfort? Forego vacations that involve any significant travel?

The political "solutions" to this don't actually resolve the calculation of the cost. They don't actually make people decide that they'd rather have a lower CO2 footprint than heat or cool their home. Primarily they just try to hide such costs for as long as possible and, when it's finally discovered just how high the cost is, foist the blame off on somebody else.

When you start having to apply force to make people choose a certain way, that's a good indication that you're making them give up something they value more in order to prioritize something they value less. Now maybe it really is one of the rare cases where the vocal minority is both right and totally incapable of convincing everyone else. But are you sure? If you were really sure, you'd be willing to bet your own money on it, and if you were right you'd probably win...

To sum up: It is nearly impossible in the long-run for a small group of individuals to calculate the "optimal" "social value" to pursue more accurately than society as a whole already does, and you should be wary of those who claim they wish to try because it's nearly always that they want to enforce their own, personal preferences at disproportionate expense to everyone else's happiness compared to what the would-be enforcers actually provide.

  • I think this is a good answer (+1), but there are many counter-arguments. In particular "In a free market, the amount earned is directly related to societal benefit." is a questionable claim. For example stealing or drug dealing can be profitable activities, but it's generally agreed that they don't benefit society. Imho the amount earned is only related to how much the customers want the product and how much they can pay for it, and there's no direct link between this and social benefits. There's also the obvious bias of inequalities: a few rich people can afford to pay for things ...
    – Erwan
    Feb 6 at 0:19
  • ... which increase their own happiness but can be detrimental to society as a whole. On the other side many poor people would really need things which would increase social benefits (e.g. housing, medicines,education) but they can't pay for it. There's also the bias of short-term vs. long-term, for example there's still a lot of money to make in the oil industry because people want cars now, but there's a serious serious chance that this will turn out to cause many deaths in the future. In general the point about the impossibility for an individual to comprehend the full picture applies ...
    – Erwan
    Feb 6 at 0:27
  • ... in exactly the same way to the market, which is also made of people who can't take into account everything. In particular the market cannot anticipate or quantify uniquely disruptive events like a global climate disaster, and exactly like in a bubble there is still some profit to be made by some individuals until the crash... and by doing so they make the crash even more devastating for the whole society. Anyway this discussion could fill thousands of pages, indeed.
    – Erwan
    Feb 6 at 0:41
  • Just to add a conclusion: essentially the market aligns with the benefits of those who have money to spend and in proportion to the amount they have to spend. These benefits are not necessarily the same as those of society as a whole.
    – Erwan
    Feb 6 at 0:58
  • @Erwan So you need to run your calculation through another iteration. The people who have money to spend: Where and how did they get it? In an ideal free market they got it by providing socially-desired value to other people. They are then entitled to request help with their own desires in proportion to how much they have helped others. Humans being what they are, this is really the only way you will convince them to continue serving others. Trying to skim off some of that reward to do something else always results in the people you are skimming from being less willing to work for others.
    – Perkins
    Feb 14 at 19:52

There is a simple mechanism that can help alleviate this problem: setting ranges for both minimal and maximal wages and this by sector and also by enterprise and these are reviewed in a yearly fashion. And within these limits, sectoral or enterprise collective bargaining or both.

Here, sectoral refers to a sector of the economy, say nursing or investment banking. Whilst enterprise refers to a firm, like BUPA or Goldman Sachs.

According to the Wikipedia article on maximum wage, countries with sectoral collective bargaining generally have greater union representation than those with enterprise collective bargaining. For example, France and Belgium have sectoral agreements covering over 90% of the working population whilst the US and Greece have enterprise agreements covering only around 10%. The UK also has enterprise and some sectoral agreements covering around 30%.

Such sectoral or enterprise agreements need not be subject to minimum or maximal limits, this is a separate measure.

Venezuela in 2012, set a maximal limit of 16 x minimum wage. State governors for example, could only earn 12 x.

Such a policy has a long history. One of the earliest recorded is by Simon Islip, the Archbishop of Canterbury in 1350 who condemned "priests who cared more for money than the safety of their souls" and instituted a maximum wage for them with fines if found in fault. More recently, Wesley Lloyd, a US state representative for Washington proposed a constitutional amendment that would limit earnings in one year to one million dollars. It failed to gather enough votes. Roosevelt during WWII proposed a maximum wage limit through taxation: over a certain $25k (around $400k today) earnings would be taxed at 100%. He failed to push this through, but achieved a high rate of 88% (and in fact, at time of 93%) on earnings over $200k (around $3.2M today).

Today, many green parties have such a policy commitment in their manifestoes.

  • One can invent many schemes, but does it actually work? Does France (or Venezuela ;) live better than the UK (or US)?
    – Zeus
    Feb 3 at 7:05

The most common tools are the government's spending powers and the government's taxing powers.

Governments routinely pay people to do things deemed socially valuable, either as government employees (e.g. all of the Orthodox Priests in Greece) or through government grants (e.g. New Zealand's Wizard of Christchurch who was funded mostly with a government grant). Lots of academic and scientific and medical research and educational work gets funded this way. And some of the workers about whom complaints are frequently made about their social value not being adequately compensated (e.g. teachers and nurses) are frequently in positions in which they are government employees or are funded with government grants.

The power to tax is routinely used to impose taxes on economic gains which are deemed to be excessive, and tax breaks (deductions and credits and exclusions from income) are routinely made available in connection with economic gains and conduct deemed to be valuable.

For example, in the U.S., charitable and religious organizations are tax free, donations to them are tax deductible to donors and excluded from gift and inheritance taxes, and the implicit income from providing housing in kind to clergy is not taxed. There are also credits and tax breaks available for conducting research and development, for employing people who categorically often have a difficult time finding employment, for protecting the environment with a conservation easement, and for building low income housing. Meanwhile, some kinds of activities are more heavily taxed than normal, such as selling marijuana, retaining too much profits in a corporation not taxed on a pass through basis, using plastic bags at grocery stores, consuming alcohol and tobacco, and so on.

There are also "private law" and regulatory mechanism to address people who are perceived to make too much money based upon undesirable conduct. When the benefits are at the expense of someone else, we authorize the losers to sue the winners in the relationship.

For example, spouses often do considerable unpaid work for each other and their children in the course of a marriage. But if a couple divorces, the unpaid work is compensated for by assuming a "partnership theory of marriage" that divides property and awards alimony in a way that equitably divides the gains that the couple as a whole made economically during the marriage, effectively treating the unpaid work in the marriage as definitionally equal in value to the paid work in the marriage.

Laws mandating that decedents make minimum provisions for a surviving spouse are directed towards more or less the same end.

Similarly, discrimination law seeks to redress within a firm overcompensation of some workers and under compensation of other workers relative to each other for impermissibly discriminatory reasons by allowing the workers discriminated against to sue for money damages arising from this misallocation of resources for impermissible reasons.

Collective action by unions, through collective bargaining agreements and labor action to secure collective bargaining agreements, similarly tries to secure equity between different participants in the same firm.

On the other hand, the government finds certain criminal activity to be definitionally undeserving of compensation, and in furtherance of that prohibits people who are owed money in connection with criminal activity from suing to be paid what they are owed, and furthermore makes the assets of those people subject to civil and/or criminal forfeiture, and imposes fines on these people.

which mechanisms have been proposed to actually solve this perceived problem and achieve wages that are aligned with social value production (however one may define that term)?

This question seems to contemplate a global restructuring of the economy by one primary mechanism that aligns these things, and I don't think that this has been done or even seriously proposed outside of a purely communist regime (and even then, not really, as that often simply aspires to give everyone what they need without regard to whether or not it was earned).

I don't think that this is possible, in part, because the social value of particular kinds of production isn't easily established and is deeply rooted in normative judgments particular to each particular activity.

Instead, the solution, as ugly and messy as it may be, is to identify particular groups of workers who are overpaid or underpaid relative to their social value and craft solutions on a case by case basis, in much the way that the market valuation of work is not the product of some top down overarching single determination, but instead arises from, in an economy the size of the U.S., for example, billions or trillions of individually negotiated transactions each year.

If we think parents are undercompensated for their parenting work, we could subsidize parenting through taxes (as we do), publicly provide funding for obligations of parents (as we dod in public education), and could as a matter of private law impose duties on children to generously support their children (as historically was common as strong social norm).

If we think professional athletes or college football coaches are overcompensated, we could cap their salaries, or impose taxes specific to that kind of activity.

If we think teachers are underpaid, we could increase their pay by raising taxes to pay for it.

If we think investment bankers are overcompensated, we could more heavily tax earnings from capital and property than we do.

One answer proposes a national "maximum wage" but that doesn't do much to help the problem of two occupations where most workers are paid less than the maximum wage having the wrong wages, based upon social value, relative to each other. It may mute the impact of a few people being overpaid, in much the same way that progressive income taxation does, but it doesn't fix the relative incentives to pursue one line or work rather than another.

One could certainly commission studies to better examine which workers and activities produce positive and negative economic externalities. But it would be foolish to believe that such studies are not more art than science. Ultimately, someone has to evaluate those studies, find them to be credible, and come up with a workable way to reconcile the situation with the status quo on a case by case basis. Most likely the solution should reflect the nature of the economic externalities. Ideally, if people in one occupation imparts positive externalities on another group of people, we would tax the beneficiaries for the benefit of those providing the benefits.

But this can go too far. Suppose that you could perfectly account for all externalities and transfer funds so that net of the transfers there would be no externalities, positive or negative, in the economy. This would still mean that very capable people who do a lot and earn compensation would be very highly compensated, while people who are impaired in some way would receive little or no compensation, even though they need economic support to live. Lots of people are in the little current contributions boat: children, elderly retired people, people with disabilities physical, mental, or cognitive, and the ubiquitous category of unconnected people with bad luck whose unfortunate choices (whether justified or not) has led them to contribute little to society. Unless we want extreme social Darwinism, simply paying people based upon the value that they contribute to society, when some people are more able than others, is itself unfair in a different way.


The price system is a rough way of estimating social value. It is not perfect, but so far it is the best way of communicating scarsity information* in the economic system. Negative externalities (such as environmental damage) can be factored in with government intervention (such as the carbon tax or incentives for renewable energy), positive esternalities are less of a problem because they are usually captured by the market participants one way or another.

* i.e. how much of good X is required vs how much it is demanded. The good in question can be apples or bananas as well as one's work. The more a good (or work is demanded) is demanded the more it is beneficial, bar some degenerate cases which are often addressed by governments.

  • 1
    This is literally just restating information that is already in the question.
    – user141592
    Dec 18, 2021 at 19:28

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .