Ezra Klein, There Is A Reason You Can't Get An Answer to Your Question

Ezra Klein writes:

For a long time, I took questions about stifling innovation very seriously. So did a lot of liberals. But then I realized that the people making those arguments wanted to do things like means-test Medicare, or increase cost-sharing across the system, and generally reduce costs in this or that way, which would cut innovation in exactly the same way that single-payer would hypothetically cut innovation: by reducing profits.

I also found that I couldn't get an answer to a very simple question: What level of spending on health care was optimal for innovation? Should we double spending? Triple it? Cut it by 10 percent? Simply give a larger portion of it to drug and device manufacturers? I'd be interested in a proposal meant to maximize medical innovation. I've not yet seen one.

The reason he could not get an answer to this very simple question is that it is stupid.  It is a non-sequitur.  It is, as Ayn Rand used to warn, a statist trying to force the argument to conform to his statist assumptions.

Let's take a different example, because medicine is so screwed up by government intervention that it can be confusing.  Let's imagine ourselves in the computer market in 1974.  The market is dominated by IBM mainframes, and innovation at the time was considered to be the penetration of mini computers (not to be confused with PCs, these were really just smaller mainframes) by DEC and HP.

Let's say that for some reason the US government decides it is fed up with the IBM "monopoly" and the high cost of mainframe computing and it wants to take over.  It feels like there is a lot of waste in mainframes as some people are using them for frivolous reasons while other companies who really need them can't afford them.  They might have created review boards to make sure that they thought each dollar spent on computing hardware and software was "worth it."

So, how much spending is needed to maintain innovation?  We know in hindsight that the PC revolution is looming in the next few years.  And in that context, Klein's question is absurd.  The answer is that spending per se, and even profits, in the mainframe computing market were irrelevant to the coming series of innovations.    The necessary preconditions were that entrepreneurs saw that new technology provided potential new value to consumers, and were allowed the freedom to launch these new products in hopes that the value these new products provided would be sufficiently high that consumers would pay enough for them to return their cost of manufacture and development and return them a profit.  Some succeeded, and some failed, but entrepreneurs were allowed to try, despite most "experts" predicting the PC was a silly toy.

Note that computer innovators were not required to trundle into some government computing board to justify the PC and its price, to justify how much, as Klein would say, needed to be spent on PC's.   If in fact they were forced to do so, if Jobs and Wozniak had to fly to Washington to justify the Apple I to the Computing Spending Decisions Board, they would have almost certainly been shot down.  Or told they could sell it but only for $200 and not their initial price of $2000.  We would have never had a PC revolution in a government single payer computing world, no matter how much, as Klein asks, was "spent" by the government.   It is possible that the government might eventually have greenlighted a PC (years later) just as the increasingly bureaucratic IBM did, but can you imagine how frail the PC revolution would would be if only IBM had ever sold PCs, without the slew of competitors that emerged, and if every innovation had to pass the scrutiny of a government review board before it could be launched?   Only a tiny percentage of PC innovation and of what we think of as a PC today, mostly in the basic architecture, ever came from IBM.

The very problem is that when government runs computers or health care, innovation is seen as a cost.  Klein, by asking the question in this way, is betraying exactly what is fundamentally wrong with a single-payer system.  The single-payer tends to think in terms of trying to deliver the current value proposition (ie the 2009 level of health care technology) as cheaply as possible.  The problem is that in 2039, it will still be focused on delivering the 2009 level of health care technology.  For the government -- a new drug, a new procedure, a new test -- these are all incremental costs, to be avoided.  Klein just wants a number he can plug into budget projections to say, "see, innovation is covered."  Its like Wesley Mouch asking John Galt near the end of Atlas Shrugged to tell him what orders to give.

I wrote about it just the other day.  You can see it in everything the Left writes -- increased spending is equated with increased costs which are therefore bad.  They all say that America's health care spending is rising and our per capita spending is higher than other nations and that this rising spending is somehow a problem to be fixed.  But there is a value side of the equation.  What are we getting from the spending?  When you leave out things the health care system can't do anything about (homicides and fatal accidents) Americans have the longest life expectancy in the world.  We are getting something for that extra money.  It is not just "cost" to be contained.  Is a year of life worth an extra $100,000 spending?  Everyone has a different answer, which is why we typically let each individual make these tradeoffs, and why people are uncomfortable having someone in the Post Office make the tradeoff for them.

But, the left will say, we will put really smart people on this board, who are angels of public service, who will make perfect decisions on the price-value tradeoffs of innovation (have you noticed that all their programs seem dependent on this assumption?)  Back to our computer example, these guys, they would argue, would have been smart enough to have given Jobs and Wozniak the green light.  This is a fantasy.  It never happens.  No matter how good the people, every such government entity is driven by its incentives, and this group's incentives will be to cut spending.  Innovations that result in a net total increase in spending are not going to be well-received.

Further, these boards get politicized, always.  Companies will quickly learn they have a better chance, say, of getting a new breast cancer treatment rather than a new prostrate cancer treatment past the board because the current administration is closely tied to women's groups.  Just look at current government R&D spending, this already happens.  AIDS was under-funded given its mortality because Conservative administrations thought it a disease mainly of groups it found distasteful; today, women's cancers get far more funding than men's due to the strong political activism of women's groups and the success of the pink ribbon campaign.  Drug companies will learn that the quickest way to board approval may not be winning over the board, but getting certain interest groups to lobby the board, or maybe lobby Congress to override the board.  Just look at the promise not to politicize ownership of GM -- that lasted about 2 days before Congress was passing legislation reversing internal GM decisions and GM was making plant closures based on political rather than economic concerns.

But even beyond these problems, there are Hayekian ones as well.  In the mid-seventies, there might have been only a few thousand people who were excited enough to buy an early microcomputer and see its potential.  What are the odds that one of those folks would be on the government review board, particularly since few of them were in the mainstream establishment of the computing field (heck, few of them were over 19 years old).  And even if one were on the board, would they have approved a technology with only a few initial adherents?  The fact is innovation often requires adoption of bleeding edge risk-takers who are willing to try a new technology and iron out its kinks before the mainstream catches on.   The iPod was not the first music player -- a few of us struggled for years before the iPod with large and sometimes hard to use early mp3 players  -- but if these early MP3 players had not existed, the iPod would not exist.

Perhaps most importantly, everyone makes different tradeoffs.  It may make perfect sense for some person in Washington that a biopsy is not required for certain kind of positive cancer test results.  This may make perfect price-value sense to the beauracrat, but I know a number of people who would lose months or years of their life to worry -- worry that could be short-circuited with an inexpensive biopsy.   Or consider a new cancer treatment -- is a year of life worth an extra $100,000 spending?  Would I prefer to extend my life through chemo or increase the quality of life of the time I have left by avoiding chemo?   Everyone has a different answer, which is why we typically let each individual make these tradeoffs, and why people are uncomfortable having someone in the Post Office make the decision for them.

One could say that all of this does not answer Klein's question.  That is because his question, built on the wrong premise, is unanswerable.  I suspect he knows this and is, as Brad Warbiany posited in the link above, just setting up a straw man.  All I can do is try to give a feel what what innovation does require, and help folks to understand that it has little if anything to do with Klein's question.

So, if I had to come up with a pithy one sentence answer, here it would be:

Klein:  What level of spending on health care is optimal for innovation?

Me:  The very fact that you intend to control spending centrally, at any level high or low, is what kills innovation.

Postscript: For a totally different reason, I was reading this article on the Russian T-34 tank, probably the best all-around tank for its time ever made when considering its production volume (the Panther was theoretically a better tank but volume production of the scale of the T-34, not to mention mechanical reliability, eluded the Germans).  Apropos of government boards and innovation was this:

The L-11 gun did not live up to expectations, so the Grabin design bureau at Gorky Factory No. 92 designed a superior F-34 76.2 mm gun. No bureaucrat would approve production, but Gorky and KhPZ started producing the gun anyway; official permission only came from Stalin's State Defense Committee after troops in the field sent back praise for the gun's performance.


  1. DrTorch:

    Your computer analogy is an apt one. Largely b/c it went unregulated for so long, and it is fresh on people's minds.

    Here's another take in that direction...say gov't boards had been appointed. Say they took innovation very seriously, and wanted to foster developments. What would they have done in 1974?

    Probably they would have looked way ahead to the future and chosen to finance more innovation along the lines of the Cray I-A

    That's what "smart" people do. Cray was a genius and his computer developments revolutionary. Surely this would have been a no-brainer.

    However, today the top supercomputers are "beowulf" systems: systems where thousands of PC-processors are linked together, forming affordable, scalable, adaptable systems.

    Solutions come in a variety of forms. I won't say the free-market is "perfect" at finding solutions, but I can say that a "panel of experts" is a guarantee of not finding many attainable solutions.

  2. MAS1916:

    This is a great piece! The absence of substantive response to legitimate questions won't help anyone win an argument.

    I think you've also hit a real point here that relates to the increased level of animosity in our public discourse. Pelosi said yesterday that she was now afraid that someone would resort to violence when they didn't agree with the Democrat's agenda. ( This is after she called them Nazis. )

    Part of what fuels the animosity as you point out - is lack of any kind of substantive answers to legitimate questions surrounding cost, rationing and service levels to citizens. Platitudes instead of quantitative responses only cause further contentiousness. ( for more on why there is such a level of contentiousness, you can view: http://www.conservativeblog.thewebinfocenter.com/conservative-blog/pelosi-concern-over-public-discourse )

    Combining that with Obama's bullying of opponents has driven support for health care takeover to its lowest level. If Obama comes out on Sunday to intimidate the opposition, look for the animosity to increase yet again.

  3. pdb:

    As soon as you mentioned government managing computer development, the sad tale of Japan's Fifth generation computer popped into my head: http://en.wikipedia.org/wiki/Fifth_generation_computer

    The Japanese put together a government run consortium of industry and academia, burned through billions of dollars in R&D over ten years, did everything exactly wrong and never even produced a sellable product.

    That's what happens when you throw government money at a problem. Sometimes I wonder if the reason we don't have a cure for cancer is that government is funding it. Why, if they solve it, the funding goes away!

  4. nicole:

    Great post. The alarm bells go off for me as soon as I read something like "should we double spending?" Who is this "we" you speak of? The right level of medical spending is: whatever each individual decides he wants to spend.

  5. James:

    God damn this is well written. I wish somebody like the NY Times would pick this up. Kudos

  6. ElamBend:

    Honestly, that first paragraph of Kleins that you posted is amazingly illogical.

  7. J Howe:

    Great response. You just need to fix 'prostrate' cancer with 'prostate' cancer.

  8. Greg:

    Good criticisms of a truly socialized medical system, but I'm not sure universal health insurance is such a threat to innovation. Can you demonstrate that Medicare has stifled or displaced medical innovation for those over the age of 65?

  9. txjim:

    A Coyote Opus! Right on brother. Political influence on technology is guaranteed to wreck anything and everything that makes a normal person's life easier. I spent a few years back in the 90's working with medical app comm technology (HL7). At the time there were several hundred Mom & Pop specialists (little dogs) that helped big dog dev shops manage app integration. It was a time marked by frequent innovation and yes - disruption.

    The big dog's took over the standards definition board then lined up to lock out the little dogs. Little dogs were mostly gobbled up and hired by the big dogs and life moved on. Predictably, innovation slowed to a crawl and the focus turned from code tuning to the machinations of the bureaucracy.

    The lesson I learned: If the technology is simple you can bet your ass the certification will be expensive, painful and indecipherable.

  10. Evil Red Scandi:

    I'll just continue to pile on the praise for this piece (just remember that if you hurt yourself patting yourself on the back, Obama won't cover it yet). Klein's remarks remind me of this great quote from Herbert Garrison: "There are no stupid questions, only stupid people."

  11. Quincy:

    Greg, the plans for universal insurance currently in play absolutely threaten innovation. The incentive set imposed by the Democrats creates a system in which insurance companies are on the hook for massive payouts with few outlets to limit costs. Not paying for new treatments is one of the precious few ways these companies could stay afloat, thus removing the incentive for developments of new treatments.

  12. Bearster:

    Coyote, great essay. One thing I would add is that Klein is committing the fallacy of begging the question. When one begs the question, one's question presumes what one should be asking.

    When did you stop beating your wife?

    How much should government spend on health care?

    There is no answer, because the proper question is: should government take money from healthy or rich people to spend on those who are unhealthy and poor? NO!

    Check out this speech at a recent Tea Party in Arizona, from a doctor who has an uncommon insight into why there won't be quality (much less innovation) under socialized medicine:

  13. Dan Maloney:

    Love the piece!
    Having grown up in the computer world since the IBM PC and all its successors I can easily see the analogy and I agree completely.
    Good thinking!

  14. Dr. T:

    "When you leave out things the health care system can’t do anything about (homicides and fatal accidents) Americans have the longest life expectancy in the world."

    Not only that, but our elderly are healthier and more active, on average, than those of other nations. Living to an old age while still being active is worth lots of money.

  15. Douglas Foss:

    I wonder how disproportionately Americans spent per capita on computing in the 60's and 70's. What a completely wasteful approach. The numbers undoubtedly would have demonstrated it.

  16. ilovebenefits:

    I am not a lawyer so someone will likely find the loophole in what I am about to say, but here goes. Wasn't the Sherman anti-trust act created to do exactly this...spur innovation. While at its core it was to protect monopolies from taking over a market and driving up costs to consumers, doesn't it in fact enable innovators to enter the market?

    I suppose, in its infinite wisdom the Federal Government is exempt from the anti-trust laws.


  17. Bob Sykes:

    I hate to disabuse Mr Klein of a fantasy, but Medicare Part B is means tested, as I found out a year ago. A lump sum payment I received on retirement was counted by SSA are regular, ongoing income and my Part B payment was increased and deducted from my SS benefit.

  18. Link:

    Oracle (relational databases), SAP (enterprise management), EMC (storage), and of course Microsoft-Intel (PCs) were all businesses that originated in some way at IBM. IBM failed to capture the new markets these companies came to dominate because it didn't have perfect foresight.

    Our telecom revolution is in large part built on the success of unswitched TCP/IP technology, much of which was developed by the propeller hat crowd at the big telcos. But the likes of AT&T would have kept us in a switched world if they'd had their druthers. If it weren't for competition our daughters would think that getting a princess phone extension of their own was the height of cool.

    If world class profit-driven companies can't get this right, how can we expect a government bureaucracy?

  19. Val:

    Greg, perhaps you should ponder why, when there is such a huge and growing market for it, geriatric medicince is woefully underpopulated by geriatric practitioners. Hmmm...

  20. stan:

    A really great article on how govt bureaucracies create incentives that are exactly the opposite of those in the private sector. http://www.eternityroad.info/index.php/weblog/screeds/14/

  21. Maximum Liberty:

    It seems that Ezra Klein completely missed the socialist calculation argument that Hayek won. The question of "how much should 'we' invest in XYZ industry" -- based on innovation or any other criteria -- is not something that government can ever get right in the sense of reflecting all information that everyone has and reflecting everyone's preferences.

    Beyond that, the idea of anyone needing to submit a proposal to Ezra Kelin to have their investment decisions authorized is offensive.


  22. Chip:

    I agree with your coments, we need to point out that the vast majority of innovation in medicine is for the American free market. If we increase goverment involvment what medical wonders will we miss. PS I think your example of aids spending is 180degrees off. Aids was a virus affecting a tiny subculture. In s very short time the gay civilrights movement made it the cause celb and spending on it vastly out paced other more wide spread illnesses.

  23. rsm:

    Fantastic piece, now if they'd included that in my degree program I might actually have found some ****ing benefit there.

    One thing that a lot of pundits like to forget is that 'extraordinary claims require extraordinary evidence' (paraphrasing Micheal Shermer et. al.) - and it should apply equally to psychics and government.

    The free market, for better or worse, works. Claiming that the government can do something *better* than the current setup requires proof. Not wishful thinking and polemics, but actual data supported with evidence and carefully analyzed under current relevant standards. Now, as always, it is up to the one who makes a claim to prove its validity. The claim is not that reforms are necessary. They may very well be, as there are always ways to improve competition and efficiency in most markets. The government is claiming that their intervention will somehow improve things, except they don't have any metrics expressed other than cutting costs and increasing insurance coverage (to include people who don't want/need insurance). Right.


    Aids was first identified in that subculture, and was then identified with that subculture. At which point it became known as a 'gay disease' or a disease related to IV narcotics usage, horribly stigmatizing etc. This is magical, stupid, wishful thinking and contrary to heterosexual wishful thinking AIDS was and is an equal opportunity blood borne/sexually transmitted infection (Oh look, one of the first AIDS associated deaths was of (a man, his wife and their nine-year old daughter - he got it sailing on the African coast) and although it may have beat out real killers like diarrhea and malaria for funding it still has managed to decimate Sub-Saharan Africa, helped happily along by a bunch of lunatic and malicious quacks, magical thinking and ignorant politicians. Try reading wikipeda for some basic AIDS/HIV historical information, and current information on its level of seriousness (mostly only beat by malaria), rather than repeating ignorant assertions. And to refute your basic assertion: AIDS was not, when it was discovered, a virus ONLY affecting a tiny subculture. The only possible exception to this would be in the very beginning of the spread of HIV-1 to humans, before its identification, when it was most likely only carried by a couple of members of the bushmeat hunters subculture in the Congo.

  24. epobirs:


    It's a bit more complicated than that. There are major differences in AIDS transmission due to genetic differences in the populations involved. The indications are that past plagues that devastated Europe favored a subset with favorable genes for resistance that came to dominate the European gene pool and consequently became dominant in North America and other places settled by Europeans. This offers an important clue for researchers fighting the disease. This has been seen elsewhere in history. Smallpox was still a dread disease to Europeans but as the product of many hundreds of generation who'd live with it among them, they had it far better than the populations in the Americas who had their numbers massively reduced by the virus when those Europeans brought it with them.

    It works both ways. There are horrific diseases that pop up occasionally in Africa but tend to be limited outbreaks. Populations outside Africa are likely far more vulnerable. If Ebola somehow found its way to a major US urban center the results could be beyond anything seen since, well, the Black Plague.

    The greater resistance of those descended from survivors of European plagues is the reason AIDS was first found among specific groups whose practices massively increased the probability of infection. Namely, IV drug users who shared needles and sexually active gay men. Resistance won't save you if you force the issue. Recipients of blood transfusions were another subset tied directly to the first two and easily reduced to nearly nothing once there were effective screening techniques.

    Given a dozen generations, the population of Africa would be dominated by those less prone to infection by AIDS but that is hardly a satisfactory answer to the problem.

  25. Corky Boyd:

    "But, the left will say, we will put really smart people on this board, who are angels of public service, who will make perfect decisions on the price-value tradeoffs of innovation (have you noticed that all their programs seem dependent on this assumption?) Back to our computer example, these guys, they would argue, would have been smart enough to have given Jobs and Wozniak the green light. This is a fantasy. It never happens."

    This is precisely why the PC revolution didn't start in Europe where governments and government boards pick a favored company (usually very large and government connnected) to manage new programs. The PC revolution happened because private capital (often from venture capital firms) and bright minds made it happen. Often the breakthroughs came from small firms with small budgets, and often large outlays came up short. Government boards abhor acknowledging failure and generally try to play it safe.

    Your analysis and comparison to health care is is great.

  26. Mesa Econoguy:

    Most of Klein’s stupidity has been covered already here, but a couple things:

    1. Klein seems to think that cost reduction via means-testing and other ways is similar or equivalent to single-payer economic structure. It’s not. Single payer will destroy profits and eliminate innovating firms altogether.

    2. Klein is hubristic enough to seek a “proposal” to quantify innovation, an unquantifiable (and unpredictable) metric, as coyote brilliantly illustrates above. This guy is Genuinely Stupid ™.

    Again, I ask who is this Ezra Klein, and why is he writing about economics? He clearly has almost no knowledge of the subject matter.

    This debate (along with many others) is boiling down to an argument between those who understand economics, and those who don’t.

  27. D. Saul Weiner:

    And of course Klein has never considered that sometimes we spending a lot of money in order to obtain optimal value in health care is exactly the wrong thing to do. Sometimes the free market would favor simpler, more basic treatments which are sound and cost-effective. These treatments tend to get short shrift in our current system where state-approved MD's favor expensive and high-tech treatments which are paid for by 3rd parties.

  28. peter:

    I have been told that those on the cutting edge are the ones that usually get cut. It is this fact alone that would prevent the government from ever being on the cutting edge. They would never let themselves get cut. While there may be some chance that innovation could occur with government involvement it would be on an evolutionary basis not revolutionary, which means it wont be seen in our lifetimes and probably not for many generations to come. For that matter god help the doctor who is accidentally innovative. How long will it be before they are allowed to practice medicine again. They will probably spend the rest of their lives explaining to the government and appologizing for saving a live in an unapproved manner.

  29. epobirs:

    This makes a nice companion piece: http://www.latimes.com/business/la-fi-hiltzik21-2009sep21,0,5599248.column

    The LA Times Michael Hiltzik, in addition to be a sockpuppeting jerk, has an incredibly bad track record for being wrong about just about everything. Here he tries to claim that the internet could only come about by whim of the government, and tries to carry this rationale into a basis for government being required to deliver health care to the masses.

    He conveniently skips over the fact that this is only the case because of prior government interference constraining private sector innovation. Almost thirty years after the AT&T breakup, he still hasn't noticed how innovation in the world of telephony and related applications exploded. He mentions that AT&T stood in the way of early networking but neglects to mention that AT&T only had this power because the government had granted it a monopoly.

    We need a new term to describe Stockholm Syndrome in a relationship between government and citizen.

  30. Mark:

    It is really interesting to add the T-34 example to this argument. One thing that is missing in this argument is that innovation is not the only important factor in "innovation". The T-34 may have been the best armoured vehicle of WWII, but would it have been a viable, commercial innovation? I dont think so.

    Maybe when your back is against the wall and you have essentially slave labor working for you, and cooperation (sometimes stolen) from the West, you can mass produce a fine weapon. But could the Russians translate this "innovation" into anything else?

    Further, the claim about the "governemnt success" in the Internet is idiotic. The reason why the Internet is valuable is because it is so decentralized. If individuals and innovators would not have created the content the Internet would just be the internal email system the government intended it to be.

  31. TDK:

    One of the problems that government funded schemes suffer from, which you do not address (not that it harms the post), is that secondary considerations start to impact the primary objectives.

    For example, the UK government used Nationalised industries as a means to alleviate unemployment blackspots. It opened Steel plants in such areas to cut down unemployment. Needless to say, the lack of natural reasons to site the plants (qualified staff, infrastructure, materials etc) didn't help and these types of enterprises usually failed

    A more recent example is the growth in non-jobs. The government perceives that it needs to address inequality, so the NHS is used to implement PC policies. Posts are created with the function that they will enforce equality and stop isms like racism and sexism. A new bureaucratic layer is being employed in many places to ensure compliance with sustainability. These people cannot in themselves increase productivity so you need a vast amount of lost production due to racism, sexism etc to justify them. I have my doubts about that.

  32. Another guy named Dan:

    Historical example of the phenomenon you're addressing. In the early 20th century, the US government created a joint military/civilian/university research project to create a heavier than air flying machine. Under the auspices of the Smithsonian Institution (then the primary US government science agency) and with funding from the Army, one Professor Langley staged some rather spectacularly unsuccessful experiments off of a barge in the Potomac.

    Meanwhile, two bicycle mechanics from Ohio, working in a shed in North Carolina, succeeded.