Fertile Fallacies and Policy Bubbles

Recognising when previously useful ideas are causing harm

These days George Soros is probably best known as the subject of demented conspiracy theories that have crawled from the far-right recesses of the internet into the rhetoric of populist politicians around the globe. Unfortunately for Soros he hits the conspiracist jackpot: Jewish, a financier, and an extremely generous funder of liberal causes. In 2022 he gave the Democrats considerably more money than will be spent by every party combined in this year’s UK elections.

Despite becoming a political punchline, Soros’ own writings have had relatively little attention outside of the finance world. They are, however,  full of interesting ideas with implications well beyond the bond markets. In a series of lectures he gave in 2009, after he had, again, been proved right about a looming financial crash he explained his theory of market bubbles:

“Every bubble has two components: an underlying trend that prevails in reality and a misconception relating to that trend. A boom-bust process is set in motion when a trend and a misconception positively reinforce each other. The process is liable to be tested by negative feedback along the way. If the trend is strong enough to survive the test, both the trend and the misconception will be further reinforced. Eventually, market expectations become so far removed from reality that people are forced to recognize that a misconception is involved. A twilight period ensues during which doubts grow, and more people lose faith, but the prevailing trend is sustained by inertia….Eventually a point is reached when the trend is reversed; it then becomes self-reinforcing in the opposite direction.”

The critical point here is that bubbles are usually based on real trends. The dotcom crash in 2000 was built on the very real arrival of the internet, and the 2008 crash on a genuine increase in the cost of housing. Soros develops this into the concept of “fertile fallacies”:

“We are capable of acquiring knowledge, but we can never have enough knowledge to allow us to base all our decisions on knowledge. It follows that if a piece of knowledge has proved useful we are liable to over-exploit it and extend it to areas where it no longer applies, so that it becomes a fallacy.”

It occurred to me that this idea could have a useful application in thinking about policy. We have a tendency to talk about policy ideas as if they are intrinsically good or bad when in reality they are often a fertile fallacy – a useful idea that has positive outcomes and is then over-exploited. To illustrate the point here are three different examples which hopefully illustrate its wide applicability.

Fertile Fallacy 1: Education is the best way to reduce inequality

Back in the 1980s and 1990s there were a lot of schools with intakes from low-income families where almost no students left with good exam results. Not in every instance, of course, but “well what do you expect with these kids” was a fairly common belief amongst policy-making elites. In 1985 just 40% of students stayed in full time education beyond 16, and these were predominantly from wealthier households.

Previously most of these early school leavers would go straight into employment but the deindustrialisation of the Thatcher years increasingly left students without qualifications with very limited prospects. Andrew Adonis recalls, in his memoir, a visit to a school in Sunderland, after he was appointed Tony Blair’s schools adviser, where the headteacher told him that the students would have once: “left to get a job in a shipyard or right to go down to the mines. All those jobs have gone now. They might as well walk straight on into the sea”.

For 20 years or so from the early-90s onwards there was a cross-party effort to change this mindset, with the heavy use of Ofsted and standardised exams to put pressure on schools. Under New Labour, there was also a big increase in funding. This was supported by changing expectations within the teaching profession driven, in part, by a new generation of more idealistic school leaders (though, of course, there had always been idealistic teachers doing their best in an underfunded and low-expectations system).

It's hard to assess real changes in attainment over time, because there was quite a lot of grade inflation in the 90s and 2000s, but far more low-income students now get a decent set of GCSEs with many more going on to do A-levels, or equivalents, and then university. Almost three quarters of all pupils are now in full time education post-16. Newspapers that once used “comprehensive” as a swear word spend far less time these days worrying about the perils of inner-city schools.

These changes were underpinned by a political rhetoric that claimed education was the most important element in reducing inequality. This quote is from Michael Gove in 2009 (when I was working for him) but is indicative of views expressed by not just politicians from all parties but also numerous people working in education over this period:

“[Education is] a driver of real social justice. The very best means of helping all realise their potential – of making opportunity more equal – is guaranteeing the best possible education for as many as possible.”

I didn’t write that speech but I wrote others like it and I believed it then, and continued to do so when I went to work for Teach First after leaving government. Without that idea being so widespread we would never have got permission to introduce extra “pupil premium” money for children from lower income families (subsequently watered down by a funding formula that undid the benefits).

The only problem is it isn’t true. Even as education has improved overall, the gap between rich and poor students has remained stubbornly wide. It reduced slightly up to 2018 and has since grown again. Nor has there been any change in wider UK economic inequality – indeed deep poverty is getting worse. A better education system is a good thing in itself, and, at the margins, might help with poverty reduction, but it is not by any stretch the best way to do that. As I wrote in the FT last year:

“Politicians, from all parties, love the idea that education is the answer to inequality. It’s intuitively plausible, appeals to those who see personal merit and hard work as the primary cause for differences in wealth, and avoids having to talk about the real problems. But the reality is that states can only meaningfully reduce inequality by providing substantial financial support to those who need it, through either the welfare system or labour market interventions. The most equal countries in the world are not those with the best education systems but those with more redistributive social policies….It is ludicrous to expect schools to salvage a situation in which children are going hungry and cold in overcrowded, dilapidated housing. If, as a society, we genuinely care about reducing poverty, we have some obvious levers to pull that we are choosing to leave untouched.”

Here we have a belief that was genuinely important in improving the education system over several decades, and which still adds value in providing motivation to teachers and leaders, but which also gives politicians a way out of acknowledging the importance of welfare. To push back too hard on the idea seems to risk a return to a culture of low expectations. But not to risks continuing to ignore the reality of poverty. It perfectly encapsulates the challenge of fertile fallacies in policy.

Fertile Fallacy 2: Governments shouldn’t try to “pick winners”

In the 1960s and 1970s government direction of the wider economy was very much seen as a necessity. Both Conservative and Labour governments attempted complex national economic planning exercises working with business and trade unions. Active industrial policy played an important role. As the economists Diane Coyle and Adam Muhtar explain:

“Nationalisation and subsidies were the primary tools of choice from the postwar years up to then, and were used to resist deindustrialisation by supporting declining industries such as steel producer British Steel Corporation, auto maker British Leyland, and machine tool manufacturer Alfred Herbert. Attempts to create new economic engines, such as commercial nuclear power, did not lead to sustained success. These episodes left a generation of British policymakers a lasting distaste for ‘picking winners’, given how much they had ended up ‘backing losers’ instead. Experiences such as these were reinforced from 1979 on by an explicit ideological push to retrench the role of the state. Since Mrs Thatcher’s election, paving the way for the large scale privatisation, spending cutbacks, and deregulation programmes in the 1980s, the received wisdom in the Treasury and elsewhere has been that the best industrial policy is no industrial policy.”

This shift was undoubtedly better than what had gone before – pouring large sums of money into failing businesses was not a good use of scarce resources. Repeatedly doing so helped cement the widespread view that the UK was “the sick man of Europe”. Better, if bumpy, economic growth up to 2007, was seen as, in part due, to this proper embrace of the free market.

More recently, as productivity has failed to recover from the 2008 financial crisis, there has been renewed interest in industrial policy from all parties. Theresa May was the first Tory leader in a long while prepared to embrace the concept and published a strategy. Likewise willingness to talk about the state’s role in the wider economy seen as an important difference between Starmer and Reeves’ Labour versus Blair and Brown’s. The logic is that it’s possible to support key growth sectors like green energy, with tax breaks and subsidies, rather than preserving dying industries.

But the fertile fallacy – “don’t try to pick winners” – has made it hard to fully embrace the idea. It’s hardly the only barrier – the highly centralised and unstable nature of Britain’s political system is big problem – but it has proved important. As Coyle and Muhtar say:

“The Treasury’s focus on other equally important policies such as fiscal management and public finances often means that issues of industrial policy often get a lower priority, and also fall foul of the Treasury’s institutional aversion to intervention by government.”

Which doesn’t mean no one in the Treasury is keen on industrial strategy, just that its institutionally downplayed, especially as the Business Department (under whatever name it has this week) is usually viewed with wary disdain. Meanwhile, Rishi Sunak also has the fertile fallacy buzzing around his brain which means he won’t even utter the term industrial strategy (he binned May’s when he became Chancellor). Or, more importantly, let anyone else do so even when we do have policies to support certain sectors. Former Tory Business Secretary Greg Clark described the  current approach as ‘surreptitious’, adding:

“One of the points of an industrial strategy is that you should know what it is. To have an undercover strategy is self-defeating.”

Fertile Fallacy 3: Transparent fiscal rules help control spending and give markets confidence

I have been critical of our current fiscal framework in several posts. But again it is based on an important truth. In the 1970s the Treasury had far less control over spending than it does today and projections were usually overshot as departments spent more than they had said they would. After the 1976 IMF crisis, when the UK was forced into agreeing a humiliating bailout, Treasury officials finally won permission to start introducing tighter controls.

In the early 1990s, with spending again rising faster than projections, Chancellor Ken Clarke forced his cabinet colleagues into agreeing an internal rule to spend less than annual growth. Gordon Brown turned it into a public rule, albeit one based on his definition of the economic cycle and Treasury forecasts, which gave him a fair amount of wiggle room. It was Alistair Darling who turned it into a more rigidly defined rule, and then George Osborne who created the OBR to monitor it. As explained in this post – the net effect has been to push Chancellors into pretending they will spend far less in the future than they will and then spending the proceeds of this pretence.

A policy belief that initially began with an important truth – governments need to have control over state spending and some process to maintain it – has ended up distorted into an absurd farce whereby Treasury officials are frantically changing their policy proposals for the Chancellor based on daily fluctuations in projected borrowing for 2029.

Creating Soft Landings for Policy Bubbles

Once you start looking there are fertile fallacies littering the policy landscape. The overreduction in NHS hospital beds and deep reluctance to add more stems, in part, from the real benefits accrued by speeding up recovery times and reducing numbers significantly during the 80s, 90s and 2000s. We just took the idea too far.

The Fujitsu post office scandal is the latest high profile example of dire government procurement practice, and their overreliance on a handful of “too big to fail” suppliers (which in the case of Fujitsu was partly a result of the failed “picking winners” approach). But outsourcing was a good idea when it began – and it did save money on things like bin collection and cleaning, where the services contracted out were straightforward and easy to measure. Unfortunately as it became applied to more and more things that were extremely hard to measure, and thus much more difficult to contract for, things started to go wrong. But ministers across successive governments stuck to the belief that the private sector is always more efficient regardless of context. And that’s how you end with Chris Grayling attempting to outsource the probation system to utterly disastrous effect.

It's the same story with welfare. Putting some conditions on benefits payments to avoid fraud was reasonable enough, but we have ended up with destructive sanctions that push desperate people further away from the labour market. Another idea taken too far.

The advantage of seeing these things as fertile fallacies rather than straightforward errors is that it should help avoid the trend becoming “self-reinforcing in the opposite direction”. If we are unable to adjust these policies to acknowledge the misconceptions contained within them, the risk is that the eventual backlash, when the failures become unsupportable, means we lose the initial benefits. The Truss/Kwarteng attempt to completely avoid any kind of fiscal framework is an example of what that looks like.

Rather than burst bubbles we need to find soft landings, like central banks are attempting with national economies at the moment. That means rather than taking simplistic yes/no positions on issues like whether schools have a role in ending inequality; or whether we should have an industrial strategy; or a framework to control government spending; or outsourcing; or benefits conditionality – we need to identify the true benefit and the misconception. We need to split the fertile idea from the fallacy.

At the moment politicians from all parties seem scared of questioning many of these fertile fallacies for fear of being accused of wanting to reverse the gains (Labour’s position on fiscal rules and benefits being obvious examples). But, somewhat ironically, pursuing a misconception to the point the bubble bursts is the thing that will ultimately reverse those gains. Admitting the problem will save them.