11 September 2012

Is government better at R & D?

Research and development are the twin drivers of the innovation that improves our lives, pushes out the productive frontier and delivers profits to those with the foresight, luck and perseverance required to bring better mousetraps to market.

What role should government play in this process?
I drove past Solyndra's Fremont, CA plant a few weeks ago.
It was for sale, but taxpayers will not get their money back
In general, I think that government is useful in promoting (but not necessarily providing) education from kindergarten through to PhD and funding basic research. I do NOT think that government needs to be involved in developing research ideas into marketable products and processes.* For-profit companies can do that quite well -- especially when they are subject to competitive pressures.**

This preamble leads us to the question of whether government should take credit for the fracking revolution that is currently seen as a "good thing." Alex Trembath and his colleagues at the Breakthrough Institute think so, but I disagree, and here's the comment I left on their site:
I continue to see flaws in the claims that (1) "public support for fracking" made the difference in its development and (2) public R&D money is well spent.

On (1) see Philpy's comment in the article, i.e., " I think your article overstates the role that the Federal Government played in the Barnett Shale development."

On (2) one must measure success not by a single case (even that's debatable) but by comparing Gov't spending against ALL returns on that spending (while controlling for non-Federal financing and personnel).

From what I've found on the interwebs, federal funding has been found to (a) increase salaries but not effort and (b) not be more efficient than spending in other rich countries, even if it has won awards [pdf] in areas where research may have been done -- or not -- without DoE/Federal interference/crowding out.

In short, Breakthrough has provided evidence that "government investment in innovation can, over time, commercialize and deploy technologies that make yesterday's less-efficient, dirtier, and more expensive technologies obsolete," but it has provided NO evidence that "can" means "must." The government, in other words, is sometimes useful but not necessary for research. Breakthrough needs to work harder or make more realistic claims.
Bottom Line: Government should stick with activities that the private sector cannot execute, not displace private sector activities by pursuing politically popular programs (case-in-point).
* Development agencies drilling deep wells for drinking water in Bangladesh were so enthusiastic that they drilled thousands of wells before realizing that the water had harmful levels of arsenic. Now it's too late to go back and millions are being poisoned by their only water supplies.

** Businesses can screw up, as American Airlines found out when passengers used its "lifetime first class" pass to the limit. AA, sadly, tried to void their agreement rather than keep their word.

H/T to JW

3 comments:

Bradley Stark said...

Disagree. Government historically has funded development of innovations when they need to cross 'the valley of death' between invention and profitability. Look at the tech bubble wherein 90+ percent of companies imploding with private funding.

A significant percentage of new technology companies will fail. Waiting for big energy companies (fossil fuels receive $517 Billion in world wide annual subsidies and oil $371B) to fund wind and solar...good luck. Only government can do it, should do, does it well by spreading money around broadly and not 'picking winners' within an industry.

Read Michael Grundwald's book on Stimulus Bill and its excellent description of funding renewables.

Alex Trembath said...

"Government should stick with activities that the private sector cannot execute, not displace private activities by pursuing politically popular programs."

I agree with this, which is precisely why government should invest in technologies and industries with clear long-term social and environmental benefits that the private sector is unable and unequipped to pursue. Spillover effects from R&D, lack of immediate profit potential of new technologies, and incompatibility with existing energy systems are a few of the many reasons why government has a clear role in energy technology innovation. The reality is that the energy sector values stability over innovation, that demand for energy is highly inelastic, and that energy industries recycle less than 0.5% of revenues into R&D. What these realities show is that price signals for innovation or deployment are signals to a system that is categorically unprepared to interpret them, and that government investment is critical to fill the gap in developing technologies that might take decades to provide significant commercial value.

There was no market for microchips when the Air Force guaranteed the purchase of all chips manufactured by Fairchild and Texas Instruments, and now we have an entire commerce/communications system dependent on them. Gas companies asked the DOE for help in funding/coordinating R&D and demonstration projects so they could effectively extract the gas in shales they knew was there but didn't have the tools to get to.

I'm sympathetic to your complaints David, and I respect your sirens for cronyism/corruption. But my challenge to you is this: if not government or public-private partnerships, then who? Please respond with a case study in which a general purpose technology was developed completely void of federal support.

Lack of evidence is not evidence of absence; obviously if there is no such case, that does not prove a priori that government MUST invest in later-stage technological development. But it will show that the alternative model has never or rarely been tested.

Interchangeable parts. National weights/measures standardization. Railroads. Airplanes and jet engines. Radios. Semiconductors and microchips. Highways. Nuclear power. Personal computers. The Internet. Shale fracking. Solar/wind/vehicle batteries. Cell phones. Carbon fiber. GPS. Gas turbines. Biotech. Robotics.

Some of these came straight out of government labs (nuclear power, GPS). Others required government infrastructure investments (railroads, highways). Some were the product of extended public-private partnerships (Internet, shale gas, robotics). All of them were improved when the private sector picked up on the platform technologies (iPhones, Internet, GPS, etc.).

Abstract arguments about rent-seeking and displacement of private investment are compelling but unconvincing, since they amount to an untested hypothesis. I would ask that you a) explain which of the government programs that developed the above technologies you would retroactively cut, and b) explain how the private sector would have developed the technology anyway, without government support.

Bradley is right to point to Grunwald's book — you might also look at Ruttan's Is War Necessary for Economic Growth?

David Zetland said...

@Alex -- I am not sure if you're defining "general purpose" to include "things that require government" but I'd start with food and transportation over thousands of years. I'd then go for guns and other weapons, engines (steam, diesel), music, art, medicine, etc.

You're RIGHT that governments have monopolies and the ability to throw money at problems (for better or worse), but you need to pay more attention to opportunity cost (how much are Iran's people paying to get the bomb? are we paying for Iraq's useless war) and crowding out (why is the government in the insurance or home finance business?)

THEN add rent seeking and corruption. Need examples of that? Read the papers.