22 March 2019

Perverse incentives in healthcare

The perverse incentives in healthcare are pervasive, worrying and unsustainable. Joseph Jarvis speaks eloquently about those that afflict the American healthcare system here. One example: he points out that investment in urgently needed new antibiotics is too low because the pharmaceutical industry prefers to develop drugs that will be taken for a lifetime, rather than a couple of weeks.

What the US and many countries have is a market not for health, but for treatment. It's nothing like a free, competitive market, being subject to the usual distortions and inefficiencies that ensue when big business manipulates the government and regulators. It's called a market, because much of it's run, nominally, by the private sector. But we are seeing worldwide a convergence of the interests of the public sector and big business, which often takes the form of policies that conflict with the goals of ordinary people. 

The 'market' is often evoked rhetorically, and that unfortunately discredits the whole notion of the market as the most efficient way we have of allocating society's scarce resources. There is a market in the US for healthcare, in the sense that the industry reacts rationally to the incentives on offer. But the incentives have little to do with the well-being of citizens, and a lot more to do with the short-term, narrowly measured, goals of doctors, insurance companies, hospitals and pharmaceutical companies. The perversity is that the healthier the citizenry, the more parlous the state of the medical industry.

Where does the Social Policy Bond idea come in? My starting point would be to define and reward the achievement of society's health goal, so that the structures and activities of the sectors that support that goal would be entirely subordinated to that goal rather than, as now, the other way round.

  On a national level, physical health could be defined as a range of targets, all of which would have to be reached and sustained before we can say we have achieved our goal, at which point Social Policy Bonds targeting health could be redeemed. My suggestion is that our goal would include such targets as: longevity, Quality Adjusted Life Years, and infant mortality. There would be others, to be decided by experts in consultation with ordinary citizens.

Where does this get us? It puts in place a system whereby people are rewarded for bringing about actual improvements in health. Not for screening, or curing or treating disease, nor for selling drugs or health insurance. Those are indirect means to an end, rather than ends in themselves and the results are lamentable: pills that are no better than placebo (see here and here). Or incentives to over-diagnose and over-treat. Or to falsify or otherwise manipulate the results of drug trials.

Health Bonds would change all that. All the activity they stimulate and reward would be entirely subordinate to society's health goal. There would be a market - for the bonds - but it would be society's servant, not its master. The Social Policy Bond principle uses the market as a means to society's goals. It doesn't view the market as an end in itself. Under a Health Bond regime, the the end that the market serves under a bond regime will be society's health, as defined and targeted by society itself. The goals of those working in any field impinging on society's health would be exactly congruent with those of ordinary citizens: to improve society's health as quickly and efficiently as possible.

16 March 2019

Link to essay on Environmental Policy Bonds and climate change

My previous post described how we might fold the effects of climate change into goals for Environmental Policy Bonds, so that we could tackle all our urgent, big, environmental challenges, regardless of their cause, in ways that are more likely to generate buy-in than the current intense focus mainly on greenhouse gas emissions. I've written a longer essay on that theme, which is more suitable for people unfamiliar with the Social Policy Bond principle, here.

08 March 2019

Climate change and the environment

Climate change could be tackled in various ways. We could aim to throttle its causes. We could aim to keep the climate, as measured by a wide range of physical, biological and financial indicators stable. Or we could aim instead to target for improvement all aspects of the environment, whether or not their degradation has been caused by climate change.

Policymakers have mostly gone for the first option: to try to reduce greenhouse gas emissions, believing that they are the main driver of climate change. Myself, I've advocated in the past the targeting of an array of indicators, including the impacts of adverse climatic events, but also physical, social and financial measures of climate change.

I'm now more inclined to the third option, and think we should try to solve our environmental problems however they are caused, rather than focus on trying to prevent climate change. My thinking is partly a result of humanity's having done almost nothing actually to stop the climate changing.

 For buy-in we need meaningful outcomes


What’s wrong with targeting greenhouse gas emissions?

  • We don’t really know what’s happening to the climate;
  • We don’t really know why it's happening;
  • We don’t know whether trying to reduce greenhouse gas emissions will solve whatever the problem might be;
  • We do know that targeting greenhouse gas emissions will have large upfront costs, and that any benefits will be way into the future, uncertain and, even on the best advice of the experts, tiny.

Admittedly the science appears to say that greenhouse gas emissions are changing the climate in ways that adversely affect human, plant and animal life, though it’s less convincing about the effects of reducing these emissions. But that’s not really relevant. There’ll be no action taken, and indeed there have been no significant actions taken (see here, here and here), because, in my view, the relationship between greenhouse gas emissions and adverse impacts is too tenuous and abstract to generate buy-in. And buy-in is what we urgently need.

Note that I am not suggesting we don’t target greenhouse gas emissions: only that any decision to do so needs to be made on the basis of whether it’s the most efficient way of achieving our environmental goals.  Professor Jem Bendell paints a cataclysmic picture:
The evidence before us suggests that we are set for disruptive and uncontrollable levels of climate change, bringing starvation, destruction, migration, disease, and war…
>My suggestion is that planetary well-being would be better enhanced by aiming explicitly to reduce such scourges -  starvation, destruction, disease and war, than by targeting, or pretending to target, greenhouse gas emissions. Reduced starvation, disease and war are goals that are less abstract and more meaningful to ordinary people than climate change. By targeting them we’re more likely to generate the buy-in that is essential to bring about changes that will create, at least in the short term, losers. All the evidence tells us that, after many years of exhortation, dire warnings, and extreme climatic events, there’s very little buy-in to the goal of reducing greenhouse gas emissions.

As well as aiming to reduce starvation, disease and war, we could explicitly target also environmental goals that mean something to the non-experts whose buy-in is necessary. These could include reduced air and water pollution, less noise, reduced impacts of adverse climatic events, and reduced loss of biodiversity.

It’s not so difficult: we re-frame and, if necessary, re-orientate policy to meet these goals, rather than a target level of greenhouse gas emissions which is, at best, only a putative means of achieving some of them.

Environmental Policy Bonds

I propose that we reward the sustained achievement of our environmental goals. Further, I propose that we do so in ways that channel market forces – the most efficient way we yet know of allocating society’s scarce resources – into the achievement of our environmental goals. I further suggest that we do this is by nationally or globally backed Environmental Policy Bonds. These could target our biggest environmental challenges, regardless of their supposed source. A bond regime would allow us to target long-term goals, and stimulate research into diverse approaches to solving our problems.

Importantly, the bonds wouldn't dictate how our environmental goals shall be reached. It's quite possible that investors in the bonds will find that targeting greenhouse gas emissions for reduction is the most cost-effective way of solving some of our environmental problems. The crucial distinction between such targeting, and the way emissions are being targeted today, is that bondholders will target emissions only if doing so - at the time and in the areas they decide to do so - is the best way of achieving our goals. They will be making their decisions on the basis of the science (and economics) of the relevant time and place, rather on the fossilised science of today. Their solutions, unlike today's non-solution, will be diverse and adaptive. And - another important difference - the people who look for and implement them will be rewarded only if they are actually successful in improving our environment.

05 March 2019

The metrics of tyranny

In our complex, populous societies we're not going to escape the use of metrics as indicators of social well-being. Social Policy Bonds aim to target targets that are meaningful to ordinary people. They therefore need metrics that are carefully devised, robust, verifiable and, preferably, easy and cheap to monitor. A Social Policy Bond regime, ideally, would use reliable metrics in a considered, coherent manner. There are dangers in taking metrics as ends in themselves, in using them incoherently, and in ways that conflict with people's well-being. Unfortunately, that's the direction in which we're moving. Having read The tyranny of metrics by Jerry Muller I've written about the limitations of metrics. I've also mentioned Campbell's Law. An article in the Economist does much to justify Mr Muller's and my skepticism:
Take the World Bank’s annual comparison of business regulations around the world. One country stood out in its latest ranking: China, which had languished in 78th place the previous year, jumped to 46th. India seemed to have improved, too, rising 23 spots, to 77th. Those remarkable ascents have less to do with the ease of doing business in those places than with their governments’ determination to achieve good grades. Some 40 people work in a Chinese government unit dedicated to improving its World Bank score; perhaps 200 toil in India’s. At least 60 countries have teams that focus on the index. (My emphasis)  Life and society are increasingly governed by numbers, 'The Economist', 23 February
Our societies aren't going to return to the times when policy is made for groups of 150 (see Dunbar's number). It follows that metrics will be the means to determining how well society is doing. Currently our governments rely on a motley array of narrow, short-term, Mickey Mouse micro-targets, including the de facto target of Gross Domestic Product, with their many flaws, some of great consequence. A Social Policy Bond regime, by contrast, would channel people's goals and expertise into answering fundamental questions: what should we target? What are the essential elements of social and environmental well-being? Where do we, as a society, want to be heading? Our obscurantist political systems allow our rulers to duck these questions and distract us all with spurious arguments about ideology, personality, image and sound bites.

The tyranny of metrics metrics of tyranny

It gets worse. In some societies metrics are already explicitly weaponised (from the same article):
In China, for example, Zhima Credit, a popular private service, measures “personal characteristics”, “online behaviour” and “interpersonal relationships”, among other things. A high rating entitles people to a fast-track visa for Singapore.