It’s back to the future time again. Boris Johnson is trying to wrap himself in the cape of Franklin Roosevelt and his famous New Deal, while his consigliere, Cummings, wants to go back to the 1950s and reboot Britain by building an imitation of the Pentagon’s Advanced Research Projects Agency (Arpa – later Darpa, with the D standing for defence).
Both projects have touching aspects of romanticism, ignorance and absurdity. In relation to Johnson’s FDR tribute band, his proposed £5bn splash on infrastructure comes to about 0.2% of GDP, whereas Roosevelt’s New Deal was estimated to be worth 40% of US national income in 1929. Roosevelt built dams, housing, roads and bridges across America. He restored the banking system, set up the Securities and Exchange Commission, encouraged trade unions. From 1933, his public works administration built the Lincoln Tunnel in New York, the Grand Coulee Dam and completed the Hoover Dam. Roosevelt instituted a minimum wage and maximum hours in certain businesses and asserted the right of workers to organise. For his part, Johnson will be refurbishing schools and repairing bridges – both good in their way, but on a minuscule scale. And he doesn’t seem to have any plans for reining in the City or for encouraging workers’ rights. So his Roosevelt rhetoric is basically back-of-the-envelope hogwash.
What, then, of Cummings, back from his victory tour of the north-east? Staff in Downing Street were reportedly puzzled by one of his slogans when Johnson brought him into the government: “Brexit first, then Arpa”. The Advanced Research Projects Agency was created in 1958 as part of the Eisenhower administration’s panic-stricken response to the discovery that the Soviet Union was far more advanced than the US in space technology. Its mission for the past six decades has been to “Make pivotal investments in breakthrough technologies for national security and through its investment to catalyse the development of new capabilities that give the nation technology-based options for preventing – and creating – technological surprise.”
Darpa’s $3.5bn budget is less than 3% of the Pentagon’s, but over the decades it has punched way above its weight – despite the fact that it has no laboratories or long-term employees. (Maximum tenure is five years.) It has about 100 programme managers and a professional staff of 220, so it’s small and agile – which presumably is why Cummings salivates when he thinks of it.
Its agility is enhanced by its modus operandi – brilliantly explained by Ben Reinhardt in a long essay, “Why Does Darpa Work?”. It’s not bureaucratic in the traditional sense: programme managers are given absolute autonomy to manage their projects. They report only to the director, and have almost complete authority over the teams they assemble and the money they allocate to them. And of course, they are not “lifers”.
Over the years, Darpa has spurred the development of innovations that were not exclusively military in orientation – weather satellites, the internet, GPS, modern robotics, autonomous cars and voice interfaces, to list but a few. However, one should not forget that Darpa works only for the US Department of Defense and sticks close to its original mission. This is why attempts to “clone” it have usually proved disappointing. For example, in 2007 the US Department of Energy tried to create its own version, Arpa-E – which doubtless has done good work, but has never emulated Darpa’s range of technological advances.
Cummings is not the first government official who has dreamed of capturing this magical fount of innovation. And he won’t be the last. But two problems will have to be solved if a UK Arpa is to succeed. The first is that there will have to be absolute clarity about what its mission is: who exactly does it work for? And the second is that a way will have to be found to give its programme managers the kind of autonomy in the spending of public money that their Darpa counterparts enjoy. In this country, the dark shadow of the National Audit Office looms over every official’s shoulder.
The great merit of Darpa/Arpa was that it managed to avoid that second-guessing. Bob Taylor, the programme manager who funded the Arpanet project (the precursor to the internet), once told a great story about how he made it happen. “In February of 1966,” he recalled, “I went into the director of Arpa’s [Charles Herzfeld] office and told him about the idea. And he liked it right away and after about 20 minutes he took a million dollars out of someone else’s budget (I don’t know whose to this day) and put it in my budget and said: ‘Great. Get Started!’”
Imagine what Sir Humphrey would have said about that.
What I’ve been reading
Trust in tech
Digital Technology and the Resurrection of Trust. Terrific House of Lords select committee report.
Hancock’s app
The rise and fall of Hancock’s homegrown tracing app. Wonderful (sometimes hilarious) Financial Times account of the fiasco.
American woes
What’s Wrong With America? Sobering Der Spiegel interview with the economists Anne Case and Angus Deaton on their book Deaths of Despair and the Future of Capitalism
