Permit me skepticism about the CHIPS and Science Act — “CHIPS” being the political acronym for “Creating Helpful Incentives to Produce Semiconductors” — to the tune of $52 billion in helpful incentives.
The cheerleader for this has been Senate Majority Leader Chuck Schumer, D-NY. “The idea of investing in companies and in science had taken a back seat for the last 30 or 40 years, in part because Democrats didn’t want to help companies and Republicans said they didn’t want industrial policy, just be laissez faire,” Schumer told the Washington Post. “It took a long time to convince people that this was so important and so necessary.”
The argument for the CHIPS Act is bound up in the competition with China. We don’t want the Chinese to get the best of us. Semiconductors, the brain cells of computers, were invented in the United States, and used to be made here, but are now mostly made in Taiwan. That’s not exactly China, but sort of.
Further, semiconductors have been a nagging part of the supply-chain problem. Also, there is a national-security argument, because advanced semiconductors are needed for advanced weapons — and we don’t want those to be dependent on China, or Taiwan, or anybody.
If the CHIPS Act were limited to military uses the downside would be less. However, the new law is based on the much wider worry that America has lost its technical edge, and that the federal investment needs to sharpen it. Sen. Maria Cantwell, D-Wash, chairwoman of the Senate Commerce Committee, said, “These investments will be going a long way in reversing the decline in federal R&D that has dropped threefold since 1978.”
Federal R&D (research and development) spending has indeed dropped since 1978, but that was a high point. Back then, most of the nation’s R&D was funded by the government; since the 1980s most of it has been funded by corporations — and corporate R&D is more narrowly focused on economic benefit. (The federal government still funds more basic science than corporations do, but its share shrunk from 52 percent in 2010 to 41 percent in 2019.)
Including corporate spending, World Bank figures show that total U.S. R&D as a share of gross domestic product — 3.45 percent — has risen every single year in the past decade and is at the highest point in 30 years. R&D’s share of GDP in the United States is slightly higher than in Germany (3.14 percent) or Japan (3.26 percent), and well above China (2.4 percent). It is substantially behind only Israel (5.44 percent) and South Korea (4.81 percent).
The CHIPS Act reaches into the marketplace and picks one industry as a winner – semiconductors. Many of the incentives are focused on manufacturing rather than strictly on research. The act aims to entice companies to build chip factories here. And to the delight of the Biden administration, one of those companies, Micron Technology, based in Boise, Idaho, quickly said it will build a big semiconductor plant near Syracuse in New York, which happens to be Chuck Schumer’s home state. If things go right, the plant will increase the U.S. share of the global semiconductor market from 2 percent today to 10 percent in several years. The company said that it would not have made this decision without the CHIPS Act. (There are also generous subsidies from the state of New York.)
Will the federal government spend its $52 billion wisely? Maybe, though government is not subject to the same incentives and constraints as private investors. People who risk their own money have an incentive to focus on profit and loss. Not that they always do; being human, they are also subject to fads. But if they screw up enough times, they are out of the game. Over time, the private sector tends to concentrate capital in the hands of people who make good decisions. The incentives facing politicians are different. Politicians want to look good to voters, and are more likely to be swayed by talk of crises and the urgency to take bold action.
The rush to save the world’s most advanced semiconductors recalls the fight in the late 1960s and early 1970s over the Supersonic Transport. The SST was a project to build a passenger airplane to fly as fast as the new fighter jets. It was announced in the summer of 1963 by President John F. Kennedy, who had also backed the project to land on the moon. The SST project was to be supported by the federal government in partnership with the private sector. To build the plane, the federal government eventually chose the Boeing Company.
Boeing engineers in Seattle designed a sleek, needle-nosed, delta-wing plane to fly at three times the speed of sound. It was a fabulous design, fitting for the Space Age.
People thought: Why shouldn’t America build it? We were the world’s aviation leader. We had invented the airplane. We had won World War II with airplanes. Always we had the newest technology. Were we going to cede our supremacy to the British and French — or to the Russians? They were building SSTs. They knew the SST was the future. It was said that the countries that took up the challenge would own that future, and the ones that didn’t would be left out.
During the 1960s, the Federal Aviation Administration estimated that by 1990 there would be a global market for 500 SSTs — a daunting number that might be enough for one manufacturer to make a profit. Obviously that manufacturer had to be American. If not, we would be left on the ground, looking up at a European sky, just as we had looked up in 1957 at the moving pinpoint of the Russian Sputnik.
But the SST had a problem. It wasn’t that the technology didn’t work; it did work. The economics didn’t work. The planes cost too much to build and operate, and the problem of sonic booms limited where they could fly. In 1971, Congress decided to kill the American SST. The British and French governments plunged ahead with theirs. They built and sold 14 Concordes for commercial use, mainly flying rich people and celebrities across the Atlantic. In 2000, one of the Concordes crashed, and the rest were pulled out of service.
The Russians also built an SST, but nobody cared about that.
In our recent history, the SST was just one of many economic enthusiasms to seize the public mind. From time to time, people became infected with the idea that this or that industry — steel, automobiles, beet sugar, shipyards — has to be saved, or that a new industry has to be started right now, at whatever cost. In 2006 the state of Washington went through an enthusiasm for biodiesel. Assured of state support, investors rushed to Hoquiam to build the nation’s second-largest plant to make diesel fuel out of canola oil, mustard oil, palm oil – all of them sustainable. The economics, however, were not sustainable. The owners struggled for several years, then sold out to a company that sold out to Chevron. The plant is still there, but the urgency has long since expired.
At about the same time as the biodiesel enthusiasm, Seattle fanned itself into a heat over the Seattle Monorail Project. Its promoters assured us that a monorail from West Seattle to Ballard would pay for itself. The history of public transit suggested that this was highly unlikely, but belief ran hot. The voters said yes several times, and a couple hundred million was “invested” before voters pulled the chain on it.
I’m not arguing that all such enthusiasms lead to disappointment. City Light was created by a political enthusiasm more than a century ago, and it turned out alright. Maybe the CHIPS Act will work out fine. You never know. Experience, however, has shown that it is usually better to let the fate of industries be decided by buyers and sellers in the market than by people elected to public office.
Your last sentence says it all, Bruce. I’d personally much rather see our government invest more R&D dollars in healthcare. That’s an industry where the financial incentives favor treatments over cures and making profit over meeting needs. Last year Pfizer spent considerably more on advertising than it did on research.
Your last sentence is a zinger, too: “Last year Pfizer spent considerably more on advertising than it did on research.” That’s not what Pfizer’s annual report shows. The report, which is available on line, shows:r
$13.7 billion in research and development costs, which includes new drug candidates in late-stage development for hemophilia, meningoccal infection, respiratory syncytial virus infection, dilated cardiomyopathy, alopecial areata, non-muscle-invasive bladder cancer, Duchenne muscular dystrophy, multiple myeloma, as well as an adult vaccine for COVID-19.
$12.7 billion in selling, informational and administrative expenses. This includes TV advertising on such products as Eliquis and Chantix, and promotion of Paxlovid for COVID. But it also includes the company’s costs for employee salaries and benefits, shipping and handling costs for its products, and other administrative expenses. So it’s a lot more than advertising.
The internet is great for quickly settling debates on facts, but of course the issue now has become the multitude of sources and, as Kellyanne Conway famously said; an alternative set of facts. My reading is more dated than yours (2020), but shows Pfizer spent $12 billion on sales and marketing, compared to $9 billion on R&D. See here: https://www.csrxp.org/icymi-new-study-finds-big-pharma-spent-more-on-sales-and-marketing-than-rd-during-pandemic/#:~:text=Pfizer%20spent%20%2412%20billion%20on,to%20%249%20billion%20on%20R%26D. As you;ll see, it isn’t just Pfizer. The study shows that brand name drug manufacturers including AbbVie, Pfizer, Novartis, GlaxoSmithKline, Sanofi, Bayer and Johnson & Johnson all spent more on marketing and selling their products in 2020 than they did on developing new treatments:
I should have also said that you were right, and that your point was well taken. I wrote from memory and shouldn’t have said ‘last year’ and should have said ‘sales & marketing’ vs advertising’. It’s difficult to write and even more difficult to write well (as you do). In contrast, it’s easy to comment.
I don’t get the analogy to the SST. It was an iconic ambition, but no one really needed SSTs.
Who thinks we don’t need semiconductors? Who thinks we can rely on the supply of chips we need, if we just leave it to the profit motive of private industry?