Think: meaningful outcomes

In our large and complex societies policymakers are, inevitably, using numerical indicators in an effort to target well-being. But these indicators are often deeply flawed. My previous post pointed out some of the flaws in the use of GDP per capita as a measure of social well-being. But there is also a proliferation of useless micro-targets, as this excerpt from the UK's Daily Mail makes clear: 

How offences can vanish 

High-profile and politically sensitive crimes, such as robbery and burglary, are reclassified. For example, a robbery may be transformed into ‘other theft’ or a burglary called criminal damage. Shockingly [sic], some offences are recorded as ‘no crime’ because there is no direct evidence. A mobile phone owner may not be able to prove it was stolen so it is written down as ‘lost’. Frontline police representatives suspect many victims do not bother to report crimes because their local police station is closed. Others no longer insure household goods and therefore do not report losses. Daily Mail, 16 May

A Social Policy Bond regime would require us to be clear about what we actually want to achieve. We would formulate policy in terms of meaningful, verifiable outcomes, rather than vague, vapid sentiment, or, as in this example, targets that can easily be gamed or manipulated. This would be inherent in the way the bond regime operates. The benefits of expressing policy goals in such terms would be equally clear: more transparency, stability, efficiency, public participation and, therefore, public buy-in.

Economic growth is not a valid goal

In April the British Government's new Chief Scientific Adviser, Sir Mark Walport, set out his priorities his first major public speech since taking office. His 'five key themes' for scientific advice in government are:

1. Ensuring that scientific knowledge translates to economic growth;
2. Strengthening infrastructure resilience for the engineered world of transport, energy, the built environment and telecommunications and also the natural world;
3. Underpinning policy with evidence;
4. Harnessing science for emergencies; and
5. Providing advocacy and leadership for science. Source

It's the first that causes me most concern (as it does George Monbiot). It reflects and amplifies the widespread view that economic growth is an end in itself. But economic growth, especially as measured by Gross Domestic Product (or GDP per capita), is not an end in itself. It is an indicator of economic activity. As a measure of well-being it is deeply flawed. It does not distinguish between helpful and harmful economic activity. It puts no value on any activity that bypasses the monetary economy. So it ignores leisure time, the environment, crime, health, and other things that are meaningful to natural persons. Crucially too, it ignores how the economic output it purports to measure is distributed within society.

The more than minimal fraud is in measuring social progress all but exclusively by the volume of producer-influenced production, the increase in GDP. J K Galbraith, 'The Economics of Innocent Fraud', Penguin Books, 2004.

This identification of societal well-being has permeated the thinking of our politicians, officials and now, it seems, our top scientists. We are just not in the habit of formulating policy goals in terms of outcomes that are meaningful to ordinary people. So, by default (or by conscious fraud), GDP per capita has become the de facto indicator of social well-being. We need to think urgently about changing this.
 

The dog's breakfast that is Kyoto

George Monbiot laments our failure to cut back greenhouse gas emissions:

The European Emissions Trading Scheme, which was supposed to have capped our consumption, is now, for practical purposes, dead. International climate talks have stalled; governments such as ours now seem quietly to be unpicking their domestic commitments. Practical measures to prevent the growth of global emissions are, by comparison to the scale of the challenge, almost non-existent. Via dolorosa, George Monbiot, 10 May

Attempts to cut back greenhouse gas emissions were always doomed to fail. And, in the form that they have taken, they deserve to. Why? Several reasons:

The causal relationship between emissions and climate is too obscure, scientifically and (largely) hence politically. Nobody's going to take serious action when the relationship between cause and effect is so difficult to pin down. All the costs of emission cutbacks are upfront. All the supposed benefits are uncertain and, if they ever do arise, it won't be until decades into future. At a time when our scientific knowledge of the existence, causes and effects of climate change is expanding rapidly, Kyoto and the attendant nonsensical emission trading schemes rely on science that was fossilized in the 1990s. Perhaps it was all an elaborate conspiracy designed specifically to do distract us while allowing the continued exploration for and exploitation of fossil fuels. A few windmills and higher electricity bills notwithstanding, that's basically the sum achievement of Kyoto and the millions bureaucrat-days that have been spent on the climate change issue.

A Climate Stability Bond regime would have been different. For a start, it would not assume that climate change is happening; it would not assume that if it is happening it's caused my man; and it would not have as its goal the cutting back greenhouse gas emissions. Instead it would start by specifying what exactly are our climate-related goals, all of which would have to be achieved before taxpayers lose a penny. Our climate goals would include physical, biological and financial measures of the world's climate and its impacts, all of which would have to fit into specified ranges in, say, the years 2030, 2040 and 2075, before the bonds would be redeemed. The bonds would stimulate diverse, adaptive approaches, that would stimulate and continuously respond to our rapidly growing knowledge of the climate. Despite the very long term goals of a bond regime, people would still be rewarded along the way, by doing what they can to achieve our climate goals and so benefiting from the consequent increase in the market value of their bonds.

Incentives matter. The incentives under the current system are, as Mr Monbiot has discovered, to misdirect the public and scour the planet for fossil fuels. Climate Stability Bonds would instead reward people for doing whatever we can to prevent climate change and its depredations. Because the targeted outcomes of a bond regime would be meaningful to ordinary people, they would generate participation in, and buy-in to, the approaches adopted by investors in the bonds.

The contrast with the current dog's breakfast of failed regimes is total.

Cyprus

James Meek reporting, from Cyprus, quotes Panikos Demetriou, holder of an account at a Cypriot Bank:

'We have 56 MPs,’ he said. ‘Forty of them are solicitors. Everything that goes on in Cyprus is with their consent. If they didn’t want the tax dodgers and the laundered money, they would have done something about it years ago. I’ve been here seven years and I’ve yet to see a tax dodger or anyone from the stock exchange come up before a judge so we can say: “This is the man, he’s behind bars.” Not one person. Nobody gets punished in Cyprus. Nobody gets punished and the same thing is going to happen this time round. At the end of the day they punish the ordinary person.’James Meek, the Depositor Haircut, 'London Review of Books', 9 May

It's for the best that lawyers engaged in legal proceedings should be pre-occupied with process, rather than outcomes. But making policy is - or should be - different from conducting legal cases or making laws. Perhaps because outcomes play only a rhetorical role in our policymaking, lawyers typically make up a large proportion of policymakers, and debate about policy revolves around things that matter to lawyers: process, structures, funding arrangements.

...Or precedent; a process that was satirised by Francis Cornford in Cambridge a century ago in Microcosmographia Academica: 

The Principle of the Dangerous Precedent is that you should not now do an admittedly right action for fear you, or your equally timid successors, should not have the courage to do right in some future case, which, ex hypothesi, is essentially different, but superficially resembles the present one.  Every public action which is not customary, either is wrong, or, if it is right, is a dangerous precedent. It follows that nothing should ever be done for the first time. (Quoted here, in relation to Oxford University, by John Kay.)

Tried, tested and failed will win every time under these circumstances. Ticking boxes becomes a substitute for innovation, adaptation and diverse approaches.

Social Policy Bonds could re-orientate policymaking so that outcomes would play the central role: outcomes that are meaningful to ordinary people, so that we could, if we wished, participate in the policymaking process. Or, at least, we could understand it, so that it could not be hijacked, Cyprus-style, to serve the interests of a corruptible elite.

Policymaking and journalism: it doesn't matter if you're wrong.

John R MacArthur writes:

What’s the use of being right, in journalism or politics? I gave a lot of thought to this question during the tenth anniversary of the American–British invasion of Iraq, and I’ve come to the conclusion that being right is not much use at all, at least as far as career advancement goes. No reward for being right on Iraq, John R MacArthur, 'Harper's Magazine, 18 April

Incentives are important, and they needn't be monetary. Unfortunately, our current political system, beset as it is by its own complexities as well as those of society, does not reward success or punish failure consistently enough to filter out stupid policies. Mr MacArthur quotes Scott Ritter: "Everybody who lied about the [Iraq] war got rewarded because they played the game." Exactly so. When it comes to looking back at evaluating policies - something that's rarely done - few people are rewarded in their lifetime for being right. Amongst politicians loyalty counts for far more.

Social Policy Bonds would change that. They reward people not for who they are, what they say or for whom they support, but for achieving society's explicit goals. Society's limited resources would be channelled into the achievement of these goals, transparently and impartially. Politicians couldn't get away with insane, disastrous policies; instead they would, under a bond regime, be limited to what they do best: articulating society's goals and raising the revenue for their achievement. Efficient approaches would be rewarded by the way the market for the bonds works. Inefficient approaches would receive little funding and be terminated - something that rarely happens under current policymaking systems. Under a bond regime, successful achievement of society's goals is the top priority. Under the current system it hardly features at all, as Mr MacArthur's poignant article makes very clear.

Obesity is like poverty or climate change: there's no obvious cause

Everyone thinks he or she understands obesity. Believe it or not, this is one of the harder medical conditions to comprehend. Why? Obesity is a combination of several factors: physics, biochemistry, endocrinology, neuroscience, psychology, sociology, and environmental health, all rolled up into one problem. Robert H Lustig, Fat chance, December 2012

Obesity is much like any social or environmental problem. There's very little point in government picking a particular theory or approach and backing it with taxpayers' funds. There's no obvious causal relationship. The best government, or any interest group can do, is to stimulate diverse approaches that adapt over time to changing circumstances and our expanding scientific knowledge.

It wasn't always like that. In health care it's now widely accepted and has been clearly shown that, for example, decent sanitation is essential for good physical health. Society and the environment were, generally, less complex in those days. But there is a lot more complexity now. Government would do better, when dealing with very complex concerns, to target outcomes, rather than the supposed means of achieving them.

Social Policy Bonds allow government to do this in ways that channel market forces into socially useful directions. When it comes to obesity, government should think clearly about what it wants to achieve, then reward those approaches that most efficiently achieve it. It might be that obesity itself is not a problem, but rather a marker of other health problems, as Dr Lustig says in his book. In that case, government should directly target those health problems for solution.

A Social Policy Bond regime would not prejudge the ways of tackling complex health problems. It would focus entirely on society's health goals and put in place a system of cascading incentives rewarding those who achieve them in the most cost-effective ways. For more information, check out my book or other papers, which can be freely downloaded from my Social Policy Bonds website.

It's all too complicated

Our financial system has become as complicated as our social and physical environments. All are too complicated and too rapidly changing for ordinary people ever to understand fully so, by default, hugely important policy decisions are made by people who are at best, under-informed or, at worst heavily influenced by the campaign funding contributions of vested interest groups. We aren't going to see a reduction in complexity; not in our social and physical environments. There are just too many variables and time-lags, combined with rapidly changing circumstances and growing knowledge of scientific relationships. Politicians can readily escape or deflect censure even for their worst policies, because the relationships between cause and effect are too delayed or obscure to identify unambiguously. The incentive to get long-term policy  right is correspondingly diluted.


Social Policy Bonds don't require a prior knowledge of cause and effect. Instead, they start from broad, desirable social and environmental outcomes. Say we want to increase society's physical well being, defined as some agreed, weighted combination of objective indicators, such as longevity, Quality Adjusted Life Years, etc. It would not then be up to policymakers to establish, rule on (or obscure), say, the relationship between burning coal and lung disease, or between a new drug and its net benefit; instead our politicians would, in effect, contract out the achievement of society's health goal to a motivated coalition of investors, with powerful incentives to reward the most efficient ways of improving society's health and, crucially, to terminate the least efficient.

Health, like a robust financial system and a benign physical and social environment, is too complicated for anybody to understand. But, rather than resign ourselves to cynicism and despair, we could adopt some or all of the most salient features of a Social Policy Bond regime: a focus on desirable outcomes in the form of broad, quantifiable, goals; and the channelling of market incentives into their achievement.

'Renewable' is not an end in itself

The Economist writes about the problems of burning biomass (mainly wood and crop residues) as a source of renewable energy in Europe: 

[T]he ideal of a biomass plantation that is harvested only at the rate at which it grows back is not always met. Even when it is, such plantations displace other ecosystems that would themselves have sucked down carbon. Processing and transporting the wood to the place where it is burned requires energy that may well come from non-renewable sources. [S]ome biomass programmes could end up emitting more carbon than the fossil fuels they are being subsidised to replace.The underlying problem is the reverence accorded to renewable energy itself. Greens like it for various reasons: independence from fluctuating fuel prices, rural employment, sustainability, as well as low carbon emissions. But as the sorry state of biomass shows, not all renewable-energy technologies are good at achieving all those aims. Nor are all those aims worth spending scarce public money on. Bonfire of the subsidies, the 'Economist', dated 6 April

This is what happens when instead of targeting climate change itself we focus instead on trying to limit greenhouse gas emissions; or rather, those gases thought in the 1990s to be the causes of climate change. If we want to achieve environmental outcomes, I think we'd do better to target these outcomes directly, rather than have government - heavily influenced as it is by interest groups - identify the ways of achieving them. A congenial climate, along with many other environmental and social goals, is too complex to be reduced to a set of simple invariant relationships of the sort that government can identify and encourage or discourage.

Social Policy Bonds would instead identify desirable outcomes and reward people for achieving them, however they do so. A bond regime would see investors automatically adapt to changing circumstances and our ever-expanding knowledge about the relevant scientific relationships. If burning biomass were to create more environmental problems than it solves, investors in Climate Stability Bonds would soon focus on better ways of achieving society's climate target. It would see 'renewable' and other high-sounding adjectives as possible means towards society's ends, rather than as ends in themselves.

Impartiality and health care

Chrystia Freeland writes:

[T]he muscle of the philanthro-capitalists is such that they can sometimes unintentionally distort the social safety nets of entire nations. That has been a complaint in some African countries, where the richly funded, relentlessly focused Gates programs on AIDS medicine and tuberculosis and malaria vaccines have lured local doctors and nurses away from providing desperately needed, but less glamorous, everyday care. Chrystia Freeland, Plutocrats: the rise of the new global super-rich and the fall of everyone else (page 91), October 2012

Which is why it's best to target social and environmental outcomes that are as broad as possible. Unfortunately, even under the, presumably, less idiosyncratic government-run systems of health care, expenditure is heavily influenced by groups of medical specialists with little incentive or capacity to see improvements in the general health of the nation as an objective. As a result, funding of these specialities depends to a great and varying extent, on the strength of their lobby groups or on their public profile, rather than on what would best meet the needs of society. At one time the British national health care system’s terminal-care budget: 95 percent of this is allocated to the 25 percent of the UK’s population who die from cancer, and just 5 percent to the 75 percent who die from all other causes. (Source: Alternative endings, ‘Radio Times’ (UK), 13 July 2002. This was the subject of a British Channel 4 television documentary Death: you’re better off with cancer, broadcast on 16 July 2002.)

A Social Policy Bond regime would target broad indicators of healthcare; probably including measures of longevity and Quality Adjusted Life Years. Apart from overcoming the biases of wealthy individuals or interested agencies, including government bodies, a bond regime would also see that the rational allocation of resources would not be undermined by high-profile events. Another UK example: in the aftermath of a tragic rail disaster in London that resulted in the deaths of 40 people the UK Government came under considerable pressure to order the installation of an automatic braking system for trains that go through red signals. Cold calculations showed that this would cost around $21 million for each life that the system could be expected to save. This is around five times the figure that the UK Treasury used as its benchmark valuation of a human life, which means that if the government had succumbed to pressure to install the automatic braking system it would have diverted funds from more cost-effective life-saving projects, and so caused the loss of more lives than it would have saved. A Social Policy Bond regime that had as its objective the maximising of the number of lives saved per government dollar would not waver in the face of spectacular one-off events.

Almost there

More on Social Impact Bonds in the Economist of 23 February. I've posted about SIBs and compared them to Social Policy Bonds in previous posts (here, for example). I'm glad to see the concept of payment for performance enter social policy. The flaw, though, remains that SIBs don't appear to be tradable and so, under a SIB regime, as the Economist says:

Projects which take many years to have an effect (the impact of pre-school education on university admissions, say) will not interest investors.

I have outlined my reasons for advocating tradability of the bonds here. It does appear, I'm pleased to say, that there are bodies interested in making a secondary market for SIBs.

When I first came up with the idea of Social Policy Bonds one of my colleagues told me that I was 20 years ahead of my time. He was almost right: it's been 25 years. This long pdf contains my first published paper on the subject and is dated 1988.