What are the policy implications of capital-biased technological change?

Paul Krugman has a very interesting post on this topic, so I will add a few points:

1. Taxing capital per se is not the way to go, since capital (of some kinds) bids up the wages of labor.  If one accepts Krugman’s distributional premises (not exactly my view but let’s see where it brings us), the answer is to tax mainly those forms of capital which substitute for labor, either directly or indirectly.  That would mean high taxes on the internet and other media of communications as well as high taxes on software and embedded software.  I don’t myself favor those policies, all things considered, but still I find it worth pursuing this logic to its implied conclusion.  It would imply especially high taxes on our technologically most dynamic sectors.

2. Intellectual property rights of many kinds should be weaker, as Alex discusses in his Launching the Innovation Renaissance.  That is one way to “tax” some forms of capital.  But do not expect the main action here to be found in easy-to-reproduce forms of capital, rather look to the more durable rents.

3. If you believe that the wages of labor are “stickier” than payments to capital, and there is downward pressure on wage shares, this implies a higher steady-state rate of price inflation.  I know, I know, there are various nominal vs. real finesses buried in my claim but still I think it holds up for the most part.

4. A lot of the surplus from Ricardian progress will end up captured by land, so we should follow Matt Yglesias’s recommendations in The Rent is Too Damn High.

5. A lot of the surplus from Ricardian progress will end up captured by resource owners, so we should relax constraints on resource exploration and development as an egalitarian move.  (NB: Beware tech models with only two factors!  They are usually wrong when it comes to incidence and the like; see for instance Nick Rowe.)

6. Let’s consider cutting the minimum wage.

7. Whatever we do, we should avoid mandated benefits programs — which raise the fixed cost of hiring an ever-cheapening labor — like the plague.  Whoops.

8. Lower wages strengthen the case for subbing in social security income for Medicare benefits, as the marginal value of cash is becoming higher for many of the elderly.  The case for in-kind as the appropriate form of aid is weaker.  Let’s also think about a longer-term consolidation of various aid programs into some form of guaranteed annual income.

What else can you think of?


What does cutting the minimum wage have to do with anything?

It implies the shared responsibility model, where employers pay part and government assistance pays part of a worker's maintenance. AKA the dirtball employer model.

If the wage companies have to pay workers is less, they will hire more workers, and the labor share of GDP will rise.

In theory,maybem but there haven't been very many examples of hiring decisions actually being made this way. In the 30s, for example, there was effectively no minimum wage and we still had high unemployment.

Ideally, firms make hiring decisions the same way they make capital spending decisions: will hiring more people allow me to produce more so I can increase profit. The minimum wage may be a factor for some firms, mostly in the service industry, but not that many. And the service industry is less likely to be affected by capital-bias than other industries.

The outsourcing moves weren't solely made because of the nominal wage, but also due to the differences in cost of living where you could pay for better workers with less wage than in western countries.

It's called Marginal Revolution for a reason. It's a small push in the direction that would help with the theoretical problem at hand.

Cutting the minimum wage increases unemployment because no one will tolerate unreliable dirty employees, and no one will tolerate in a political climate calling for reduced minimum wages any welfare programs to provide housing, food, and transit for the low wage workers, so they become homeless and thus unemployable.

That in turn reduces business revenue which cuts labor demand and increases the homeless and hungry reducing the numbers of employable....

People do not understand economic theory and the theory that when their income falls, they stop living in proportion; ie., a 50% reduction in income means you are supposed to live and breath only 12 hours a day.

Funny. +1

'That would mean high taxes on the internet and other media of communications as well as high taxes on software and embedded software.'

How does one tax the TCP/IP protocol (and others that tend to ride on it)? And no, I am not being cute - when someone comes over with a smartphone, uses my wireless access point to go through my router to access pictures I took on my laptop, the 'internet' was involved, as would be the transfer of data between two phones (just a reasonable assumption that TCP/IP would be involved - people who actually own a mobile phone are welcome to comment with more details).

And how does one tax GNU/Linux (for the sake of purity in terms of being completely free of any software that involves patents), or any other major piece of free software? Apache also comes immediately to mind - if you are reading this, odds are an Apache server and almost definitely BIND (BIND is an implementation of the Domain Name System (DNS) protocols - Berkeley Internet Name Domain) were involved. And this comment was written on a system that only has free software installed - which anyone is welcome to have a copy of, by the way.

The 'Internet' remains seemingly misunderstood by some - it is simply a way to transmit data between digital systems, and possible to use with only free software, without paying a penny (though admittedly, mesh frameworks always seem to die an interesting death in the face of concerted telecom opposition).

'It would imply especially high taxes on our technologically most dynamic sectors.'

Or not, as Apple's recent court cases demonstrate - patenting rounded rectangles is not a sign of technological dynamism. And software patents are pretty much an American innovation - which has led to much technological development no longer being possible in the U.S. without extremely deep pockets or patent portfolios. However, it has done something the American economy has a proven strength in - increasing the demand for lawyers.

'Let’s consider cutting the minimum wage.'

Because servants still cost too much for older citizens, one assumes.

How does one tax? What a silly question. You pass a law, then force the target to measure and remit. If you are trying to find some justification or logic in taxation, take my word for it, there are far more satisfactory endeavours for the mind.

It is very simple. Someone has money, the government wants as much of it as it can take.

I have money and I notice that the banks are trying to take a bigger slice of it than banks did in the 60s, and Wall Street wants big cuts of it, bigger than in the 60s, and the medical corporations want really bigger cuts of my money than in the 60s.

I hear conservatives who have controlled government more than anyone over the past 30 years still blaming liberals for the banks, Wall Street, and medical corporations, but conservatives have failed to deliver on their promises over and over. They have made communism look a lot better when Red China delivers what the US delivered in the 50s and 60s, and the US can't.

"2. Intellectual property rights of many kinds should be weaker, as Alex discusses in his Launching the Innovation Renaissance. " - WRONG WRONG WRONG! We've tried this for decades and it's not gone anywhere. Let's try encouraging R&D instead of discouraging it--is it any coincidence that the Solow model for economic growth only factors in technological change as the long run driver of economic growth? Why not encourage invention? Why depend on unpaid Good Samaritans (Nobel Prize winners) only for real innovation? The econ college professor's backward looking lust for 'more and cheaper of the old stuff' has led to off-shoring of American innovation--and, importantly, no long term planning for real innovation. If patent terms for pioneer inventions are only 20 years from filing,as they are today, why would a company invest billions to develop a miracle drug when, after FDC approval, it will only have a few years of life left in it? And then have to lobby for uncertain extensions? Why develop the flying car when by the time you get approval from the government it will be off patent? Google Segway. Disney gets more protection for Mickey Mouse in copyright than most companies get with truly pioneering inventions. It's no secret why PARC (Xerox) did not patent as much as they did--why bother with the present system, which encourages trolls more than real innovators?


Let's look at economic history for a moment, at a time when we did not protect intellectual property, by all agreement, very well.

Back in the 70's and early 80's federal district courts were hostile to patents. Each federal court of appeals had a different standard, which led to forum shopping. As a result, patents were weak, and enforcement was weak. Congress changed that in the early 80's by making one federal appelate court for patent appeals.

During the 70's and early 80's, federal antitrust enforcement agencies were hostile to patents and well, and practices which today we recognize as being within the scope of the patent, and thus legal.

As a consequence, during this period, US firms LOST to Japanes and Korean firms which were quite able to run around a weak patent system. Remember Xerox? Remember some of the early minicomputer companies which had good patents, but which lost to firms with access to cheaper labor and engineering in Japan and Korea.

In the mid 80's, the US woke up. We reformed our patent system and antitrust agencies issued new intellectual property guidelines. The US worked on interternational patent enforcement treaties or incorporated IP protection into treaties.

As a result, US firms could continue manufacturing here, or they could go abroad. But, they and their shareholders, ultimately, owned the intellectual property they developed with their R&D dollars.

If you want companies and countries in Southeast Asia to rip us off, you can follow this advice. Maybe the GMU's Korean affiliatecan support Alex, but you shouldn't.

I worked in software for 30 years. I guess you could say we followed a middle line on IP in various shops. We copyrighted, but did not patent, and made wide use of open source (and open framework) technologies. I kind of feel you are making an "if Microsoft had won" argument. That is, they would have given us crumbs and payed many people to applaud their innovation.

A fully owned technosphere is not really a better one.

But you make my case--Microsoft 'stole' (copyright infringed, if you believe you can copyright 'look and feel') Gary Kildall's copyright. MS-DOS is a knock off of Kildall's CPM OS. Hence Microsoft was "anti-IP"--and won. Kildall himself refused to sue MSFT since he was "anti-patent" / "anti-IP", and, says Kildall, took MSFT's word that they would not expand the market for DOS as per their IBM contract. BTW, Kildall has a case IMO, albeit somewhat weak since 'look and feel' is not widely recognized as copyrightable. But MSFT did have access to Kildall's manuals for CPM, and studied them for clues when they wrote MS-DOS. Google "copyright clean room defense" for more information (which companies routinely do nowadays, but MSFT did not).

Is it just me is are you forgetting one of the three standard impacts of granting monopoly rights on new products?

1. Static efficiency gains from competition
2. NPV drops because of lower profit
3. NPV increases because of spillovers from weaker protection

3 doesn't exist in the Dixit-Stiglitz because the varieties aren't complements in production, but we know lots of things are complements in the technology world. It also wouldn't exist with perfect contracting but I think it's pretty clear the tech world has serious contracting problems: (1) courts decide the prices for licenses in some (many?) cases and (2) patent trolls exist.

I get the sense a lot of people think the courts, transaction costs, and patent office are so bad that the NPV down effect is larger than the lost profit effect.

I think you are correct about legal transaction costs. The minimum to bring a lawsuit nowadays--in legal fees--is a couple of hundred thousand dollars. Too high. Patent Office has become a bad joke--no competition is one reason (and why they should IMO have a privatized Patent Office--that courts would look to as legal advisers, and do away with mandatory examination of all patent applications, but I digress). Re your #1: you can have competition in a patent world-- a narrowly drafted patent claim is easy to 'design around'--case in point: Amazon's "Single Click" button--simply have two clicks not one and you avoid the patent (BTW I hate the 'single click' feature at Amazon since it does not give you a chance to change your mind before purchasing, but I digress again). Re your #2: Net Present Value would decrease *for society as a whole* with weak IP--and increase for strong IP. Concrete was re-invented TWICE under a non-patent (trade secret) world: the Roman concrete (which BTW is found in nature from a certain volcanic ash) was lost, and reinvented during medieval times. If you have a legal monopoly you raise NPV for society as a whole. Case in point: NBA vs semi-pro basketball vs your neighborhood pickup game. Who plays better basketball? Clearly the NBA, since they have a "legal monopoly" on talent (as well as MLB, literally, for baseball). Same with patents--a strong IP world would 'raise the game' for innovation. I posit the present and future inventors for the (1) flying car, (2) 'cure for cancer' and (3) 'cure for old age' are working on Wall Street inventing derivatives for Goldman Sachs--that's where the money is. Or loafing in Europe doing nothing, because of the safety net there. Re your #3--not familiar with this model, but the Solow model is simple and exogenous tech is the driver--and IMO it doesn't have to be 'exogenous' or 'randomly dropping from the sky'--if you throw money at a problem it will auto-magically fix itself. Again pro sports is a good analogy.

Which of these 8 recommendatons would you rescind if we were to reject the premise that there will be capital-biased technological change?

If the answer is "none," or "only number 3," then what should I infer about how your policy recommendations relate to the phenomenon?

Tyler: i recommend this post, which looks at the steady state capital stock: http://hyperplanes.blogspot.ca/2012/12/krugman-is-wrong.html

8. Sounds like the Ryan plan. Good.

My question for both you and Krugman is why does capital-biased technological change threaten the solvency of Social Security? Benefits are indexed to the average wage near retirement. I don't know of any caps that would keep that indexing variable from declining, so the average worker pays in at a higher level than they get paid out in benefits. Sure there are some problems with this, but I don't see overwhelming solvency problems. The wage inequality example Krugman gives seems more valid...wage growth concentrated about the taxable max boost average wages, but not revenue for the program.

Another question: why does the labor's share show so much variation even as real median household income has been flat/declining over the period? I would argue there's a pretty strong cyclical pattern (with some downward trend) in the chart...seems hard to extrapolate to further sharp declines.

Lastly, what do you honestly think the political economy implications would be of "longer-term consolidation of various aid programs into some form of guaranteed"? The degrees of redistribution in our income/consumption maintenance programs vary greatly. Social Security is essentially forced savings for retirement, Medicare a buffer for health shocks (note: aging is a pretty widespread shock and generally not stigmatized.) Other programs like welfare, food stamps, unemployment benefits help buffer income against more controversial shocks. I suspect that lumping them all together would eventually leave a shell of the Social Security-Medicare programs. If that's your goal, say it plainly...you did with the minimum wage.

Isn't most "capital-biased technological change" inequality enhancing?

You mean like when we built (capital intensive) steel mills and automobile plants?

It doesn't reduce payments, but it does reduce FICA payments, because wages rise more slowly or even decline.

SS isn't forced savings. Nothing is saved/invested. It is spent on current recipients. The accounting is different, but it's meaningless. In 10 years, a good bit of the payments are likely to come from increasing government debt.

Less average wage growth does reduces new benefits, but not existing benefits. Also Social Security is saving for a household since payroll taxes paid on wage income while working translate into benefit income in retirement. Put differently what would likely happen to saving in the counterfactual world without Social Security? Finally the projected shortfall is real, but it is not 'a good bit of payments' in 10 years. See the Trustees Report link below.

Can you give an example of a form of capital that is not (or is less of) a substitute for labor?

An office building.

"Paul Krugman has a very interesting post on this topic"

If you feel this way, you must share Krugman's near-comprehensive lack of understanding of US social insurance program finance.

"Let me take it more slowly: a substantial part of our social insurance system — Social Security and the hospital insurance portion of Medicare — is funded through dedicated payroll taxes. If payrolls lag behind overall national income, this will tend to leave those programs underfunded given the way the laws are currently written."

Beginning in 2013, 3.8% of ALL income, payroll or otherwise, is subject to Medicare taxation. Now, 3.8% may not be the right number, but for this 800-lb gorilla, Krugman's characterization of Medicare finance, three days from now, will be 100% wrong.

As far as Social Security, benefits are tied to the same wage base as taxes, so, labor's share of national income will have little or no bearing on the system's intrinsic sustainability, which, AFAICT, undermines the whole premise of his article. To the extent increasing inequality has any effect, a modest ad-hoc adjustment of the wage base (It's been done before) is a simple corrective.

It is likely true that, as a percent of total retirement income, Social Security declines in the future. I would be surprised if there isn't good data on this over the past 40 or 50 years. But isn't that what we expect as a society grows richer? That Social Security, designed to provide a 'minimum floor of income', becomes relatively less prominent over time?

Bear in mind that much of the supplemental income is derived from pensions and 401(k)s, which enjoy returns to capital. Hell, the whole public sector pension system is basically counting on hefty returns on capital to pretend benefits aren't twice as expensive as they really are. But I digress.

The link between labor's share of income and the deterioration in prospects for the Social Security trust fund is imaginary. Historical trustees' reports can be found here:


Between 2001 and 2008, labor's share of income plunged, while the financial projections for Social Security were stable or improving.

All of the deterioration has come since then. The real killer for Social Security finances is wages rising slower than CPI, not inequality. And the payroll tax holiday has chipped in its share too.

"Beginning in 2013, 3.8% of ALL income, payroll or otherwise, is subject to Medicare taxation."

Not exactly. Currently, all "earned income" (wages or self-employment income) are subject to medicare tax (2.9 percent combined employer/employee). After 2013, two things will change:

1. Wages and self-employment income above the "threshold amount" ($200/$250K "modified adjusted gross income" but subject to inflation adjustment) will be subject to a 0.9 percent surcharge. Thus, the total amount for wages above the threshold is 3.8 percent);

2. Net investment income above the threshold amount is also subject to 3.8 percent medicare tax; however, the tax is on the *lesser of* the net investment income or the total amount of income (normal and investment) above the threshold amount. For example, married taxpayers with combined wages of $230K and investment income of $50K would pay the tax on $30K of their investment income.

Number two, on intellectual property, is of course correct. IP law in this country is entirely about creating a new rentier class that was lucky enough to inherit the rights to Mickey Mouse or whatever from their grandparents. Speaking of which, Disney is going to need another law in, what, twenty years or something? My guess is that they stop slow marching this thing and just go for perpetual ownership of "cultural treasures" at that point.

But the most important item on the list is number four--what should you and I do? If we are worried about the resurrection of the rentier class, we should ACQUIRE CAPITAL. As much as we can get our grubby hands on. Stocks, land, resources of every stripe. It's still a great time to swap up to a nicer home and put the old one out as a rental unit...

I don't see that Krugman made any mistake about social security solvency. He's just saying that if the wage share falls, the system is insolvent if we maintain SS benefits at his preferred level. He understands indexing full well.

"If payrolls lag behind overall national income, this will tend to leave those programs underfunded given the way the laws are currently written."

I'm all for charitable interpretations, but now 'underfunded' should be interpreted as 'inadequate according to my standards'?

I understand that you're trying to move the conversation onto what you think is more interesting terrain, but I can't let go of this. As I mentioned above, there is no excuse for Krugman's complete Medicare understanding fail.

I put on my charitable hat and reread the blog. In order to defend your view, I think we need to interpret statements such as "slashing benefits" as "not changing the law to boost benefits".

And this:

"Why, then, do most projections show the trust fund running out well before then? Not because life expectancy is rising — that was already built into the projections. No, the big reason is rising inequality"

Sorry, I do not believe the man knows what he is talking about here at all.

I understand your point on Medicare but you need to explain further on Social Security. I did not think Social Security was a "bank" where you get back out what you put in. I was under the impression that the people working today pay for the people who worked in the past. So clearly if I am correct then a decreasing labor share will imply the workers of tomorrow will not be able to pay for the workers of today. Possibly I am incorrect and (hopefully) Social Security is not a Ponzi scheme?

“I was under the impression that the people working today pay for the people who worked in the past.”

To date, the SSA had collected taxes for Social Security benefits in excess of payments to beneficiaries to the tune of $2.7 trillion (I pray that this is part of the $16 trillion debt, but I’m not sure.) This was the plan all along since 1983, to minimize “intergenerational equity” issues associated with the Baby Boom rat moving through the snake, demographically speaking.

Speaking of which, Krugman’s graph seems to me to depict precisely the long-run trend of labor’s share of income under conditions of an aging society.

Which points up the fact that every time someone retires, Krugman’s beloved ratio takes a hit.

Which suggests that this is less the romantic “labor v. capital” showdown Krugman envisions, and more of a milquetoast story.

Once the Boomer rat is through the system, Social Security can run on a more or less a ”pay as you go” basis with some minor modifications like longevity indexing, assuming some kind of steady-state long run demographics.

Provided, I hasten to add, wage increases outpace inflation, i.e. productivity growth continues. That’s really the whole ball game for SS sustainability. Introducing ‘labor’s share of income’ into the analysis is counterproductive- simpler variables tell the story.

"To date, the SSA had collected taxes for Social Security benefits in excess of payments to beneficiaries to the tune of $2.7 trillion"

"This was the plan all along since 1983, to minimize “intergenerational equity” issues associated with the Baby Boom rat moving through the snake, demographically speaking..."

This is absolutely false. There *never* was any such plan. The members of the Greenspan Commission and Congressional authors of the 1983 Reform have stated this repeatedly.

See for instance the oral history of Robert Myers, long-time chief actuary of the SSA Executive Director of the Greenspan Commission that designed the reform, at SSA.gov. Or, for that matter, the text of the Reform Act itself.

As Myers makes very candidly and explicitly clear, the terms of the reform act were set to get the members of that Congress through the next few years of expected continued poor economic performance (see how poor it was 1970s to 1982) without having to go thru the pain of raising taxes again.

Then the economy boomed, the huge SS surplus started coming in as a surprise. And Congress, instead of cutting the tax rate to the "paygo" level intended by the 1983 reform, simply spent it. Which is why Sen Moynihan, a leading member of the Greenspan Commission and Dem Senate Finance Chair, who repeatedly tried to reduce the payroll tax back to the paygo level, always called the spending of the surplus "embezzlement".

The idea that the 1983 Act **planned** to create the Trust Fund as we know it, and as it is constantly represented today, is the Greatest Urban Legend of US Politics in our time.


I don't see how this makes any difference anyway, but I offer to you page 6 from a link to the 1979 Social Security trustees report, wherein the intermediate cost estimate projects a trust fund surplus, which peaks in 2010 at 3.35 times annual payments, then declines.


The surpluses that arose when the changes were accelerated by the Greenspan Commission were not a surprise to everyone.


In theory there is a trust fund with enough money in it to pay for SS for a long time to come. Practically speaking, SS is a Ponzi scheme as the trust fund makes no difference to the Federal government's overall costs. Without the trust fund, the Federal government would have to issue more bonds, raise taxes, or print money to pay for the short fall. With the trust fund, the Federal government is going to have to issue more bonds, raise taxes, or print more money to pay for the short fall. The trust fund does not even get around the need for Congress to raise the debt limit as the "bonds" in the trust fund are special type of bond that can only be sold back the the Federal government. Thus, the Feds would have to issue new bonds to pay for the old ones and the new bonds would count towards the debt limit.

Bottom line: The trust fund is an accounting gimmick and nothing more. People try to obscure this by pretending that SS is somehow legally separate from the Federal government but this is not the case. And even if it was, it does not change the funding challenge that SS posies to the Federal deficit.

Social Security is not a Ponzi scheme! It's not a scam or black box of government waste or Krugman's vehicle for social justice... it's a well defined entitlement program that has done much to reduce the scourge of old age poverty. A broader sampling of comments: http://mediamatters.org/research/2011/09/08/social-security-is-not-a-ponzi-scheme/182872

And to reprise yesterday's guest appearance here's the excerpt in the link from Dean Baker:

"A Ponzi scheme requires ever expanding number of participants with the current participants being paid from new members of the scheme. The deal is that people are getting paid far more back than what they paid in. In fact, participants in [Social Security] get a real return that averages around 2.0 percent. It is close to being sustainable for the infinite future. The projected shortfall is equal to a bit more than 10 percent of the programs costs (by the CBO projections -- the [Social Security] trustees have a somewhat higher figure). With a relatively modest tax increase (equal to one third of the cost of the Iraq-Afghan wars or roughly 5 percent of the wage growth projected over the next 3 decades) the program would be solvent indefinitely.

Ponzis don't work this way. [E-mail to Media Matters, 9/8/11]"

There are enough problems, please don't make up pretend ones.

Brian - when you say "assuming some kind of steady-state long run demographics" that is the part I would expect Krugman is worried about. SS has been around a relatively short time. If our society becomes more stratified the projections for SS are not going add up.

Ape - I pretty much agree with your description. It is pay as you go. I used the term Ponzi since the new "investors" pay for the old.

Claudia - believe it or not I actually support SS. In my mind it is somewhat a formalization of the young paying for the old. Some would claim we should be doing this privately but it seems like we have given up on the "it takes a village" way of doing things.

Ponzi is such a wrong-headed term here. Taxes are collected for one purpose and used for another. Perhaps “shell game” is more apposite.

I think of the SS surplus as $2.7 trillion owed by the Income Tax Payers of the USA to the Social Security Beneficiaries of the USA (good news- this is part of the $16 trillion, http://www.gao.gov/special.pubs/longterm/debt/debtbasics.html, whew). These all overlapping, but in some ways, crucially distinct, groups.

More specifically, to the extent SS surpluses produced “artificially low” income tax rates, the beneficiaries of this policy have been (a) high-income taxpayers between 1988 and 1993, and (b) ALL income tax payers over the past 10 years. Figure 2 from the #3 link is useful here: http://marginalrevolution.com/marginalrevolution/2012/12/assorted-links-657.html

To me, justice demands that we get the money to pay back Social Security from these groups. (I'm in this group, currently holding a $200K reserve for my family of four as my share of the $16 trillion.)

If you take the “but it’s all one big pot of money” position, you are in essence condoning a policy of financing the general operations of the government over the past three decades disproportionately on the back of a regressive payroll tax. This position does not, IMO, put Republicans in a good light.

Oh, okay. I see how this ties to solvency of the legislated system. I was forgetting about demographics, duh. BTW I did not get this by re-reading the blog posts...scanning the summary of the 2012 Trustees Report: http://www.ssa.gov/oact/trsum/index.html helped much more. (More lines to read, but less reading between the lines needed.) Take a look at Chart B for a nice graphical summary. Lower average earnings growth are not helping the income line of Social Security.

It seems like a lot of solutions continue to lower real wages that have not increased for over a decade. At this point in time, wouldn't lowering wages continue to drop the birth rate in the US and continue to create this demographic deflation spiral?

That said a drop in the National minimum wage might move more jobs to lower wage states (ie South Carolina versus Matt Y. love of DC, NY or SF) so it might make modest improvements.

Historically the West did face these issues in the 1920 - 1930s (Ok without the Social Security solvency) but the decreasing value of wages to capital which creates a long term liquidity trap. I know the west has also avoided that solution but would a mild form of economic nationalism be a short term solution?

If one accepts Krugman’s distributional premises [...], the answer is to tax mainly those forms of capital which substitute for labor, either directly or indirectly. That would mean high taxes on the internet and other media of communications as well as high taxes on software and embedded software. I don’t myself favor those policies,

That's twisting the tax-argument and applying it selectively to items that'd get you a visceral disapproval from the audience. I don't see the internet, media nor software as more representative examples of the key capital to be taxed.

You might as well say let's tax assembly line robots or airport baggage handling conveyors but that'd get less of a reaction.

"I don’t see that Krugman made any mistake about social security solvency. He’s just saying that if the wage share falls, the system is insolvent if we maintain SS benefits at his preferred level."

This reminds me of the "lump of labor" fallacy. Perhaps we could call it the lump of GDP fallacy.

The fallacy here is that if non-wage income constitutes a greater percentage of overall GDP, total wage income must be reduced below what it would otherwise have been. It is a standard progressive meme that as wealthy persons earn more this must, necessarily, come at the expense of everyone else. This strikes me as a similar argument (and fallacy). Social Security and Medicare benefits are not indexed to GDP. While it is possible technological advances come at the expense of labor, it is also quite possible that wages can continue to grow fast enough to keep Social Security and Medicare "solvent" (under the current benefit formulae) even if "capital" represents a greater percentage of overall GDP. Stated differently, capital need not "substitute for labor", but compliment it.

And, as noted somewhere above, the new ObamaCare 3.8 percent Medicare surcharge on investment income above generally $250K will apply after Jan 1, 2013. While not all those funds go to Medicare, it does strike me that 3.8 percent tax on this income from capital is greater than the 2.9 percent tax that would apply if that income were from wages (on those earning less than $250K). As far as gross revenues are concerned, therefore, a shift towards income from capital by high-earners should bring in more, not less, revenue.

Krugman's plan (with Krugman, there always is a plan that usually involves tax increases) must be that he wants to subject the same investment income to social security tax as well as Medicare tax (without any corresponding increase in benefits).

"The fallacy here is that if non-wage income constitutes a greater percentage of overall GDP, total wage income must be reduced below what it would otherwise have been."

You are correct as long as GDP gains make up for the labor share loss. Or is there some other mechanism to increase total wage income?

Let's posit an example in which (real) GDP begins at $100 and wages constitute $50 (50 percent). Over time, real GDP increases to $200 and wages constitute 40 percent of the total ($80).

GDP gains are not "making up for the labor share loss". The "labor share loss" is a red herring. There is nothing "to make up for".

Sorry when I said GDP gains I should have said GDP gains relative to projections. So if GDP increases as predicted but the labor share decreases then we do have a shortfall. But if GDP increases quicker than expected the gain (relative to the projection) may make up for the labor share decrease.

You put far too much stock in predictions. Measuring anything against a prediction, particularly an economic prediction, is a pretty useless exercise. In fact, it is meaningless, except for what it might say about the predictor's lack of omniscience.

1. If you want to make labor more attractive, you should fund fund health, retirement, and education through general revenue instead of payroll taxes and corporate benefit programs.
a. Single payer, or at least open access to Medicaid
b. crush the rent-seeking 401k apparatus; management fees on 401ks are absurd, 2%+. The thrift savings program is <0.1%. Consider a sovereign wealth fund.
c. Universal higher education.
d. Transition: fund payroll taxes through a VAT

2. You can't directly tax labor saving capital, ** but you can eliminate intellectual property.

Perhaps we should look at Social Security differently, and remove one program from it, to be funded from general revenue.


Federal welfare support came from the federal budget. That means, the rich, if they paid more, paid a greater part of federal welfare system.

When we changed the programs, guess what happened: Some portion of the welfare population, at the advice, by the way, of local welfare agencies, applied for Social Security disability.

As a result, people who pay for welfare ala social security disability deductions, are NOT the rich...SS contributions are limited to slightly more than $110k, and are paid by all those below it.

As a consequence of "welfare reform", the middle class is paying a much greater share of welfare, and the rich are not.

I think SS disability should be taken off SS and be funded from general revenues or deficiencies should be taken from general revenues because it has become the welfare alternative.

Bill, social insurance should be funded through general revenue across the board. That basically fixes everything.

You could also start a sovereign wealth fund with carbon tax revenue.

The only problem that I see with a sovereign wealth fund is that it will invest locally, and politicians, and not the market, will pick winners and losers. I am interested in a discussion of strategic use of sovereign wealth funds that might be a bigger problem in the future.

In a sense we do social insurance backwards from your suggestion: we use social insurance proceeds to fund running deficits and tax cuts.

The Canadian one is run pretty well

The biggest one that you missed is to reduce all university professors' salaries down to the minimum wage. Let them live on this and then they would be able to comment from a more intellectual basis than the point you have set out above. If you don't have a plan other than 'lower the minimum wage' and 'don't provide any benefits' you have failed in a big way. It would behoove you and others to go out and see 'if' you could live on a minimum wage for JUST ONE MONTH. It's fine to sit there and pontificate in the absence of knowledge. At least Barbara Ehrenreich had the 'balls' to do just this.

Good Post..

Taxing capital is about leveling the tax system against the power to shift income from one to the other category. At the top we let people choose whether to consider their income is capital or labor which they naturally chose as capital to pay lower taxes. It really has nothing to do with investment other than low rates on capital promote consumption instead of investment. Mostly it implies more progressive taxation against all income from whatever source. Many of these other suggestions would be much less relevant with it, and many of your solutions are not solutions at all, but ways of compounding the problem.

Obviously, we should import tens of millions more unskilled foreigners.

To not do this would be to admit that the consensus of elite opinion over the last generation was disastrously wrong.

Why isn't population an issue here?
One of the problem appears to be that the supply of labor is too high compared to the stock of capital, making capital more productive for jobs. Then let's reduce the population to find an equilibrium. Why is taxing innovation and prosperity a more viable solution (the only one analyzed). All the humans could be living a better life with tech advances and robots by reducing the amount of work they all do, or work the same but save more for when they'll be old, it won't hurt anyone living, and the cost of execcive supply won't fall on the community.

Uh, "reduce the population" how?

What makes you think you could reduce the population and magically find an equilibrium? The distinction between capital and people is not as clear as you seem to think. Moreover, some of the problem is related to age issues not numbers of people per say as older people depend more on capital and less on income. So if you reduced population but increased the imbalance in the age structure you would make the problem worse.

Why should there be any policy implications unless one believes that central planning works and that there is some market failure implicit in the outcome resulting from the technological change?

I'm curious what market failure is occurring.

Not sure what you mean... are you saying a declining share of labor is a market failure?

How are you, personally, defining "market failure" here? I'm not restating your question, but rather asking what, for you, would count as a market failure conceptually?

@Steve, whether or not a declining share of total surplus going to labor is a market failure or not would depend on a number of additional factors.

@Meekins, my question really doesn't hang on the definition of market failure but on what justifies a market intervention via government policy.

Given this blog is largely a free-market, Public Choice view of government, I just assume that some failure is needed to justify some policy.

A much higher EITC phased out at a much higher level combined with taxing income from capital at ordinary rates after eliminating the corporate income tax and averaging real capital gains over the life of the asset.

For individuals learning in grade school or high school graduation, a totally free lunch plan is offered, especially for those of lower income homes.

The scholars who would like to obtain from college or university
federal grants should create a registration for your FASFA scheme.
t possess the funds for promote lease, even so lease
is defined by income depending across the federal poverty degree to your personal spouse and children size.

Federal government grants for single parents for houses can
be obtained when it comes to moms, and they also occur in different types.

- Be a current resident in a state or territory or tribe inside US.

A prior MR post poses the question of SS as a PS...another Krugman vs. Krugman vs Friedman and Samuelson. Some economists are of two minds on everything.


Replacing the minimum wage with guaranteed income paid for by taxing capital gains, would be my preferred approach as the least-distortionary. Taxes on capital gains have been found to have little effect on incentives and it is a way to tax the rents from capital without regard for what form it takes. In the meantime, guaranteed wages maintain the information properties of a functional economy even in the presence of disfunctional inequality.

The other thing that will be important is access to capital and direct investment. This may be another form of internet innovation (though I expect major opposition from existing banks), but more likely it will require conscious effort to broaden the base for whom capital is available. Otherwise it won't matter how loose restrictions are; the only people with access to the start up capital necessary to extract them will be the already-uber-wealthy.

Comments for this post are closed