Innovation Nation v. Warfare-Welfare State (more)

The New York Times has a lengthy piece on the expansion of the welfare state:

The government safety net was created to keep Americans from abject poverty, but the poorest households no longer receive a majority of government benefits.

…Dozens of benefits programs provided an average of $6,583 for each man, woman and child in the county in 2009, a 69 percent increase from 2000 after adjusting for inflation.

…The recent recession increased dependence on government, and stronger economic growth would reduce demand for programs like unemployment benefits. But the long-term trend is clear. Over the next 25 years, as the population ages and medical costs climb, the budget office projects that benefits programs will grow faster than any other part of government, driving the federal debt to dangerous heights.

In Launching the Innovation Renaissance (and here) I argue that the warfare-welfare state is crowding out other areas of spending, even when such spending could be highly valuable.


Just wondering, how can you establish that the decline in R&D spending is caused by the welfare state?

I thought Tyler wrote not too long ago that sudden increases in R&D spending are mostly captured by wage increases for existing knowledge workers.

Are we also going to ignore a certain something that happened in between 1960 and 1970?

>> warfare-welfare state is crowding out other areas of spending>>

There has been a huge push towards privatizing government spending, including R&D. Add in the tendency to subsidize and protect many big businesses, including through patent protection, and you get this result.

Also, crowding out is not a very effective argument in a time of massive idle capacity and close to record low interest rates.

This is why I suggest economists have stone tablets carved out that say "No evidence of crowding out."

Interesting. The CBO seems to show that R&D spending as % of GDP has been remarkably consistent for the past 50 years, around 2.5-3%. When government R&D falls, that of industry has picked up. (See the link below.)

To right-wing ideologues a rising tide lifts all boats. Even if some by meters and others by millimeters. All other scenarios are zero sum.

Definition of poverty is an effective living standard far below what most people reading this blog would consider acceptable. Demonstrably wealthy people do receive a good chunk of benefits via Medicare and SS. Let's consider addressing that before cutting Medicaid, food stamps and welfare--stuff that makes the diffetence for poor people.

A hungry kid with no healthcare or head start is far less likely to grow up and innovate.

Sounds like you agree 100% with the post's author

If there are hungry kids in America, it's because of parental neglect. All poor kids qualify for healthcare, and head start has been shown not to actually help. Any other strawmen?

There are criticisms of head start, but there is a lot of evidence it is beneficial. You can measure it many ways--your one straight refutation of its positive impact is wrong.

And just checking--nobody says Medicaid and CHIP are part of the welfare state? Pretty clearly these programs are on the chopping block and have already been extensively cut in states. Not really very straw man-ish...

What evidence are you referring to? I have never seen any evidence that it was beneficial in the long term. Even the official Head Start reports are pretty bad.

+1 from what I have seen the benefit of head start fades to 0 by the end the 3rd grade.

There are more.

Eliana Garces, Duncan Thomas, Janet Currie (September 2002). "Longer-Term Effects of Head Start". The American Economic Review 92 (4): 999–1012.

Currie J, Thomas D. (1995). "Does Head Start Make A Difference?" (PDF). American Economic Review 85 (3): 341.

This is reason why BETTER early-childhood programs should be taken up (programs which have been proven, and have proven to be very cost effective).. we shoot for the cheap program rather than the good program, and we get what (or less than) what we pay for.. whereas we'd get more with a high-quality (albeit higher cost/pupil) prog.

It is silly to lump in early-childhood programs as 'just another social program' or worse yet something the parents should be doing. Market failures abound in this case.

Demonstrably wealthy people were forced to pay money into Medicare and Social Security.

So are those programs social insurance or are they a wealth transfer? You're just making the answer more clear.

Pre-natal nutrition and (among other things) vitamin D and DHA at least to age 2.
The rest is mostly workfare for people with Master's in education who can't get a job for the Wolfers.
I'm exaggerating of course, but perhaps not in ratio terms.

The problem with going back to welfare as welfare is you broke it. In my speculative opinion, you are going to have to do a hard reboot.

Social insurance? Hmmm... I pay money into insurance every month (car, home, ad&d) and will not get it back unless disaster strikes. Why should demonstrably wealthy people expect otherwise?

Because Social Security has been sold on that basis since its inception, and would likely have been killed on multiple occasions if it hadn't been?


I thought that was a generous description of the New York Times piece. There was a lot of interesting stuff in the article, but it was hard some time to see it through the article's open cheerleading for Democrats and Big Government...

What cheer leading?

Pointing out that people suffer from cognitive dissonance means you're not rooting for the right team.

First, look at the overall presentation of the story:
--The first profiles highlight the people who seem to be the biggest hypocrites--they want smaller government but not cuts to the programs they depend on.
--Somewhere near the end of the middle, where people are least likely to read, they profile someone who provides some anecdotal evidence of waste in the program.
--The last profile is of a poor person who knows their place and votes Democratic.

Second, notice how the authors do their best to trap people with their questioning, but they don't put forth the same effort to try to understand why it might be rational for someone to oppose government spending even though they currently benefit from government programs. Given this, it's hard to imagine that the authors didn't come into the story with the intention to do what Phill talks about below--point out people's cognitive dissonance, show what dupes they are.

I'm also annoyed that what the story portrays--that there are lots of poor people who have been duped by the Republican party--is wrong. The latest political science doesn't support the notion that poor people are economically conservative. It's rich people who are. See Andrew Gelman's post here:

No reason for that graph to start at 4%.

Seriously, the y-axis shenanigans that economics and politics blogs get up to... it's really annoying, and should be called out as dishonest whenever it occurs. Does this happen in the published economics literature also?

The immediate impression of that graph is that the U.S. spends about 1/10 today what it did in 1987 (normalized to GDP) on R&D. When it's actually about 1/2 or 2/3.

There's no particular reason for the chart to bottom at zero.

Four percent could be a minimally neccessary expenditure.

"There’s no particular reason for the chart to bottom at zero."
Except for the fact that zero is the baseline that virtually every person will assume unless a different baseline is prominently pointed out.

Also, this graph would be better as real R&D spend rather than as a % of the budget. If the spend stays flat and the budget grows, there's no crowding out.

My alcohol expense as a percentage of my budget has gone down, while my travel expense has gone up. However, it's not because travel has crowded out alcohol, it's because my budget has grown a lot faster than my alcohol expense. Also, there's a limit on how much I'm willing to drink.

Wouldn't by most yardsticks of libertarian thinking warfare ( and perhaps even welfare ) rank higher in the state priority list than R&D, innovation and other such activities?

At least in warfare the state spending doesn't crowd out private spending?

For libertarians, government's main task should be to provide public goods as defined in economics (i.e., goods that are non-rival and non-excludable), such as defense, rule of law, etc. Many libertarians would argue that R&D counts as a true public good since the ideas behind an innovation, once made public, are non-rival and non-excludable.

What kind of R&D, though? I think an more likely libertarian position is - why is the government in the position of deciding what R&D to invest in when you have a perfectly serviceable market that could make far more equitable allocation decisions.

That's apparently the government opinion too as Rahul points out the market is keeping overall R&D steady as government R&D declines.

Think less about the libertarian argument, which is de facto being implemented, and more about the non-libertarian arguments.

I looked at the NYT piece, and what they included within their graphs was unemployment insurance.

Guess what happens during a recession.

The other thing that was missing was an observation that some of these programs are funded countercyclically, as you would want them to be.

That is not to say that medicare, for example, should continue the way it has, with doctors deciding what they want to give you, and not what you need. And, even for medicare, if you look at the charts, the wealthy get more from it than they put in.

And, Red States make out like bandits. Look at the Medicaid spending in the south. No wonder docs don't want managed care medicaid.

Doctors don't give you what they want. They offer, you decide. Money is only now a showstopper, why is that?

Re R&D spending graph: wouldn't it be better to show a graph of private R&D speding as a percent of GNP and public (including state colleges) R&D spending as a percent of GNP.

And, whoever said that spending correlates with R&D output.

Academics like R&D spending.

It feeds the labs and foreign graduate students.

God forbid!

They also like it because it is one of the rare true public goods.

The education that grad students get (sic) is of course not a public good, and that's a separate issue.

Oh, yeah, faculty are doing it for the public good,

And, for their jobs.

If the grad students go on to do research and teach (well), is it then a public good? (I'm a naif in this sphere...)

That's very striking. But politicians aren't able/willing to sell this. Do you think that it has to do with the length of tenure of elected officials (esp. the President) ? S/He can't waltz onto the stage and make the case that investments in R&D will lead to a better economy 20 years from now. Also there is uncertainty around the gains (probably a large spread even if the mean is positive - or may be not if you have a large number of small risky investments that are independent) and many people aren't used to thinking in probabilistic terms.

A minor stylistic point : The graph will look nicer if you use tiny url (or something else) for the link in the notes.

Spending too much on non-R&D doesn't mean R&D is getting crowded out. I need to see that absolute inflation adjusted R&D spending has decreased to believe that.

In one sense a valid point. But think about these (related) points as well :

i) To some extent the point is about the relative value of components of government spending. What is the bang-for-buck of an additional R&D dollar vs the worst non R&D program additional dollar.

ii) Government spending shouldn't be thought as being potentially unlimited. Especially now.

Treasury bills are trading at negative interest rates!

Pray tell, when else are you going to want to raise money?

There's an odd sort of parallelism between rich liberals asking government to raise their taxes and poor conservatives asking government to cut their benefits.

The main difference being the rich liberals can afford to indulge their preference, the poor conservatives are cutting their own throats.

Kind of puts the lie to the meme about the Dems using welfare to buy future votes. Seems like wherever folks receive the most government money, they vote more and more Republican.

The article in yesterday's NYT on the front page of the front section was really telling. The parts where the lower/middle income conservatives and Tea Partiers in Minnesota actually come to realize how much they rely on eeeeevil government funds are priceless. In one of their own's words, they just never thought about it before...

Back in high school we had a teacher who insisted we draw a prominant "break" symbol on the axes whenever we felt a compulsion to use a non-zero origin. Ought to be emphasised more.

It's especially egregious here because there's enough room on the graph to include 0 on the y axis.

"The government safety net was created to keep Americans from abject poverty, but the poorest households no longer receive a majority of government benefits."

There are two problems with this statement. First, the safety net exists to mitigate certain kinds of risk like, for instance, the risk that someone of Medicare age will require a $150,000 bypass operation. It doesn't make sense to restrict that benefit only to those living in "abject poverty" because few people have $150,000 laying around for such an operation.

Second, restricting benefits to only those in abject poverty effectively means taxing work very heavily for those who fall slightly above the poverty threshold. If you get a $1,000 government check when you earn $10,000 but then get nothing when your income increases to $10,500, you lose the incentive to earn more money. Things like the EITC have a very gradual phase-out precisely to avoid the possibility of effective >100% tax rates. A consequence of this is that the non-poor receive some government benefits.

I don't think the NY Times author meant that the poorest households should receive a majority of government benefits, but rather is pointing out something that many Americans are not aware of--that welfare programs today largely benefit the middle class, rather than the poor.

Isn't the marginal return on R&D spending declining?

Wonder how different that graph would look if it excluded defense and space program R&D.

Relative R&D spending might have shrunk but I suspect the portfolio is getting better.

That was my thought exactly. Much of the Sixties spending had to be devoted to the still young and popular space program (including, but not lmited to, Apollo), which spurred civilian technological development as much by accident as by design.

Any data breaking this into defense/non-defense?

A related graph has R&D Spending (Federal and Industry split) as a percent of GDP ; it seems whatever spending was cut back by the state, the slack was taken up by Industry.

So maybe all the crowding out that is happening is via federal R&D grants crowding out private R&D.

My sense (and I am not an expert on this) is that the government funds more basic research and the private sector does the R&D that takes it to market. Therefore, I'm not sure that graph necessarily suggests crowding in/substitution - it might be consistent with such an effect, but there could be other plausible explanations.

For example the Institutes of health fund basic research and the "Big Pharma" uses some of those findings to develop drugs/equipment that it sells in the market. Not sure if the first stage is viable and even it is - the positive externalities are probably huge. I think the internet and cloud computing might be another example (CERN, DARPA's ARPANET etc. did the early pioneering work and Amazon, Google and others developed these ideas further and took them to market).


The figures are very much distorted.. a consistent share (R&D/GDP) is driven by big pharmaceuticals and big auto.. the development research done by these companies may still have a sizable return for the firms that commission said R&D, but lack the public/spillover benefits that can result from basic research.

hard to have crowding out when you have so much under utilized capacity...the only thing thats restricting your thinking is the fallacy of money...

An important issue, but a useless graph.

If the federal government expands substantially because it takes on expanded responsibilities (Medicare, Medicaid), then the share of the budget for the other stuff must decline as an arithmetic necessity. The shares always add up to 100%.

More interesting question: what is the return to federal R&D? If the return is high, how much does federal R&D need to expand to push the return to the cost of funds?

In any event, it is amusing to see self-styled libertarians arguing for more federal R&D. Our schools have problems, so I guess we need more federal control of schools. Our hospitals have problems, so I guess we need more federal control of hospitals.

The return to government R&D is often very high, but because it is public, the return is spread everywhere diffusely. If a a billion of government R&D returns $10 a person,that's $3 billion return. A billion investment by Intel might earn Intel a $3 billion return and drive its stock prices to new highs.

But Intel wouldn't make the $1 billion investment without the demand generated by the government R&D in the Internet and the government policy making the Internet the winner over AOL, MSN, Sprintnet, and dozens of other private alternatives that couldn't create the boom the Internet did. Not to mention the government Y2K mandates....

Screws are cheap and quite standard thanks to government driving innovation in screws - screws as a commodity product independent of any given manufacturer.

The NYT sure knows its audience. First anecdote, Tea Party supporter gets EITC - hypocrites, those Tea Partiers are! Second, Tea Party elected official once collected unemployment - Hypocrite #2! Case closed.

Broad-based entitlements cannot be supported in the current form by taxing the rich - they do not have enough money to pay for everybody even if you confiscate all they have. So if these programs are supported by taxing everyone, why use a government program? Essentially, your money is being taken from you and you get back what the government decides is right.

Each time the Republicans have tried to rein in Medicare the Democrats have (at least until recently) scared the seniors telling them that the Repubs want to kill Grandma. A crucial opportunity to reform the program before the BBers started collecting was lost. The medicine is much tougher to swallow now, but the fantasy that there are free benefits (paid by someone else) is dying.

And each time the Dems tried to do something about health care the Reps used the exact same langauge. What's your point?

And here we are again with the either/or of cutting spending vs raising taxes. It's not either or. We have to do both and there's no logical way to see it otherwise.

If the government were doing public goods you might be right. As it is, doing primarily wealth transfers there is plenty of logical room to argue for all spending cuts.

Maybe 'logical' was the wrong word. You can logically/mathematically 'solve' the problem with all spending cuts. But in the real world of a modern democracy that simply isn't possible. So it's counterproductive to keep framing this as either/or.

Who speaks for the middle?

And the government is doing both public goods (defense, etc) and wealth transfers (Medicaid, unemployment insurance). Along with things that can be seen as either/both (Medicare). So again, we need both more revenue and less spending.

Defense is a good one. On a budget portion prorated basis do we need to increase revenue or cut defense? I'd say you can do entirely cuts in defense.

And I'd say you can't cut defense enough to make the numbers work. You need to cut other stuff too (or change benefits, like with Social Security age increases and so on). And you also still need more revenue, around 17-18% of GDP oughta do it.

Andrew, Defense is about $700 billion while the deficit this year will be around $1 trillion.

I said budget prorated basis. By that I meant that if you look at the portion of the budget that is defense, it's huge. You can actually cut defense more than most budget items. I didn't say just defense. I said defense is a good example, all the better because Republicans reflexively support it.

And private sector health insurance companies are doing something altogether different in your opinion?

Yes. They aren't borrowing 40% of their budget.

"Broad-based entitlements cannot be supported in the current form by taxing the rich – they do not have enough money to pay for everybody even if you confiscate all they have."

Nice strawman. And I hope you didn't mean to compare a stock (wealth held by the rich to be "confiscated") to a flow (annual entitlement spending and annual taxable income).

"So if these programs are supported by taxing everyone, why use a government program?"

Because they are not taxing everyone at the same levels. See pretty much any CBO analysis of the distributional effects of taxation + transfers.

Much of the welfare state amounts to a transfer from the middle class and rich to the middle class and rich. This would be fine if it were not done in a way that creates huge losses of utility.

Take education (I consider it welfare), I send my child to a private school that charges have what the county that I live in spends.

SS transfers money from my siblings and I to my parents but absent Government we would probably make other arrangements.

Medicare transfers money from my siblings and I to my parents but absent Government we would probably opt for different cheaper care and provide some care ourselves directly.

Re: "but absent Government we would probably opt for different cheaper care and provide some care ourselves directly."

Do you do heart surgery on yourself or your parents?

The other arrangements weren't taking place in the 20s and 30s because families no longer living on the farm could usefully employ their elders sufficiently to support all but the neediest and closest family members leaving many elder to county poor houses, work houses, or living in deteriorating houses with big garden plots that had been bought earlier in life. As people moved into cities as the farms mechanized, the old family safety net of living on the farm and helping out was no longer available.

By the 50s and 60s, the elderly's medical needs had overwhelmed to doctors and hospitals communist policies (each pays according to their means, and takes according to their needs). The private sector had created the private-public partnership of Blue Cross Blue Shield of prepaid medical care - technically legally not insurance. The primary way to join was through your full time employer.

"SS transfers money from my siblings and I to my parents but absent Government we would probably make other arrangements.

Medicare transfers money from my siblings and I to my parents but absent Government we would probably opt for different cheaper care and provide some care ourselves directly."

It's great that your family is so tight but what happens if your kids wind up losing their life savings in some ill-advised business venture or die young? Or what if your kids are ungrateful brats who won't give you a dime even if the government cuts you off? Even the tightest, non-wealthy families will surely gasp when a hospital hands them a $300,000 bill for grandpa's heart or brain surgery. That's the point of social insurance: to protect people and give them some sense of protection against capricious fate.

If you can afford to make your own arrangements and have a pretty secure financial situation, then you are pretty much by definition not the target of a safety net. Not everyone falls into that category, though.

"That’s the point of social insurance: to protect people and give them some sense of protection against capricious fate."

This is right, but it doesn't explain why the government has to pay for people whose children haven't cut them off, or medical care for seniors who aren't facing a $300,000 bill.

In other words, large portions of what is sold as a "safety net" really aren't.

The chart above is, without more, not enough to prove anything. I think that we would need to know what the amounts of absolute dollars that were spent, and what spending was on.

E.g. the bump in the 1960s may have been the Apollo project. It was nice, but it was essentially a cold war PR stunt, and one that won't be repeated.

That PR stunt led to some pretty powerful wealth creation in the private sector.

Sure, but here's a simple thought. As Rahul points out, total R&D spending is roughly 3% of GDP. Considering that the government is for public goods, shouldn't it's share of its own budget for one of the purest public goods be greater than the share of the overall economy devoted to quasi-public goods?

I'm not sure about 'greater than' but I'd definitely be on board with a plan that cut spending elsewhere and raised enough revenue to boost federal R&D significantly.


I don't know what the right percentages are, but I'm pretty sure that the actual public goods are getting the shaft.

Notice that every single "government" program or benefit mentioned in the article is a Federal program or benefit. No states or cities. No mention of civil organizations or associations either. It's as if either Federal government does it or nobody does it. Federal government has the monopoly on virtue.

I wasn't aware of any states or cities that had a substantial spending budget for "Research".

I might be wrong: Are there any?

Yep. The Research Triangle is a recent public private partnership that created the most recent big new center. Mass-Boston have been revitalizing the 128 miracle mile Boston technology corridor created by WWII and cold war R&D government spending. California has had ongoing huge R&D public private partnerships. Those are just the biggest longest running State-local programs to promote R&D and innovation economy activity in a State or metro area that come to mind instantly.

Well, the article was about federal taxes and the federal welfare state, so...

JFK ran on a warfare state platform and, like Eisenhower, declared innovation to win the cold war a top priority.

To connect the warfare state to the people and inspire young people to join the cold war as intellectual soldiers of innovation, JFK promised to land a man on the moon by 1970. Just like Eisenhower used civilian projects to inspire civilian innovation to win the cold war. Eisenhower invented civilian space projects to justify space flight to conceal the efforts for spy satellites and ICBMs.

Goldwater won the nomination as a result of a surge in conservative politics that attacked the civilian projects that were a key part of the Eisenhower-JFK cold war effort.

Conservatives and business allies inserted all sorts of provisions to prevent the Federal government setting up commercial civilian technology delivery or standards. This prohibited the Federal government requiring all civilian telecomm with the government use the DARPA Internet standards to allow the private sector to sell proprietary networks and lock out competitors to increase profits. It was only after 1990 that Congress removed that restriction from the Internet which was legally not used for commercial civilian purposes, although that was obviously no longer the case.

The conservatives claimed by getting government out of innovation and setting standards for commercial products to promote competition, the rate of innovation would increase as a consequence of the profit motive.

Yet the greatest driver of innovation in the past two decades have come from standards set by governments; the US government DARPA for the Internet, and the EU and various other governments for cell technology.

This repeated previous periods of innovate as a consequence of government policies defining the direction of economic innovation. Aircraft manufacture in the US was blocked by the Wright's refusing to license their patents, broken by the government forcing them to enter a cross licensing system (in Europe, the Wright patents were ignored).

Mail order commerce was spurred by Congress requiring RFD and Parcel Post - for more than a century, the private sector failed to provide useful parcel delivery for the sellers and most of the new buyers were not served or ill served by the private sector.

Commercial schedule air service was created by acts of Congress requiring the Post Office contract for air delivery from private companies that were encouraged to carry freight or passengers in addition to the air mail - the private contracts replaced the Army flights that the Post Office used previously. Of course, the Army had defined the air routes, trained pilots, and built the air fields used initially.

CNC machine tools were developed during WWII and during the cold war as means to increase US output for the wars. After the 60s, the US innovation in such technology fell as US government investment fell, while Asian private-public investment increased.

It was US government that standardized such mundane things as screws and other fasteners to simplify military logistics for field repairs especially in allied operation - the US and English manufacturers had conflicting basic designs and each manufacturer had proprietary screws each commonly used to ensure they controlled aftermarket sales.

All the upgrades of computer equipment in the 90s were driven by government policy requiring conformance (or disclosure of failure to conform) to Y2K readiness.

Comments for this post are closed