26 December 2018

Climate change: surrogate surrogate indicators

I've blogged before about surrogate indicators, or surrogate markers: 
A surrogate marker is an event or a laboratory value that researchers hope can serve as a reliable substitute for an actual disease. A common example of this is blood cholesterol levels. Evidence-based medicine in disguise: beware the surrogate, 'MD Whistleblower' (blog by Michael Kirsch), 1 August 2010
And, more broadly, in economic policy:
Instead of targeting anything as meaningful as human well-being, the de facto target of most governments is gross domestic product (or GDP per capita). 26 July 2015
But there are also surrogate surrogate indicators. In climate change, we have agreements primarily aimed at limiting emissions of greenhouse gases:
Framework Convention on Climate Change (UNFCCC), New York, 1992, including the Kyoto Protocol, 1997, and the Paris Agreement, 2015
I submit that these agreements are surrogate surrogate indicators: they are not targeting climate change; they are not even targeting the many hundreds of hard-working people who strive to achieve consensus over these agreements understandably measure success by the number of signatories to their final agreement texts.They are not, though targeting climate change. And, even assuming that greenhouse gases, as identified by current science, are responsible for climate change, the politicians and bureaucrats who, through considerable effort, get an agreement signed, have done nothing to see that even their surrogate indicator, reductions in greenhouse gas emissions, take place. Michael Le Page tells us where we are at today:
Early in the new year, if not sooner, the world will set a most unwelcome record. Global oil consumption will pass 100 million barrels per day for the first time – and keep climbing. To have any chance of limiting warming to 1.5°C, greenhouse gas emissions need to start falling now, and fast, the latest UN climate report warned in October. But emissions are still increasing. They rose 3 per cent in 2018 and look set to keep rising in 2019. Renewable energy race to ramp up as oil use skyrockets, Michael Le Page, New Scientist, 18 December (my emphasis)
And the climate itself?

Last four years are 'world's hottest'

There's a genuine problem here, despite the good intentions of the characters involved, be they scientists, politicians or bureaucrats. We have little chance of avoiding the climate disruption about which we are being warned while we are committed to the usual way in which policy is made. I don't think it's enough, when faced with a hugely complex and urgent problem, to use the science of 20 years ago as a base for a one-size-fits-all, top-down approach to its solution. We are undertaking multitudes of smaller initiatives to, for example, limit emissions and create carbon sinks. But our intense focus on greenhouse gas emissions is doing nothing to stop outrages like the destruction of the Brazilian rainforest:
Between August 2017 and July 2018, 7,900sq kms were deforested, according to preliminary figures from the environment ministry based on satellite monitoring – a 13.7% rise on the previous year and the biggest area of forest cleared since 2008. Brazil records worst annual deforestation for a decade, Dom Phillips, The Guardian, 24 November
While the goal of limiting the rise in the Earth's temperature to 1.5 degrees C is sound, there are no reasonably upfront financial incentives for people actually to achieve it. There are some, as we have seen, ineffective agreements to try to limit greenhouse gas emissions, or fossil fuel exploitation and consumption. But people react to the incentives on offer, and there are no explicit rewards to anyone contingent on the climate stabilising for a sustained period. We can take expert advice about what sort of 'stabilising' we want to see, but the important thing is that we set in place a meaningful climate stability goal and reward people for achieving it. We are not doing that right now, with the disastrous consequences that we are already seeing and more, many more, to come.

Which is why I suggest adapting the Social Policy Bond idea to deal with climate change. The essentials of Climate Stability Bonds are (1) that they reward the achievement and sustaining of a more stable climate (however defined), and (2) they will encourage the use of market forces to allocate society's scarce resources to meet our goal most efficiently. I have written more about Climate Stability Bonds here, on which page there are links to a range of my other writings on the subject.

17 December 2018

Why I go on and on about nappies

Why do I keep blogging about nappies? Because decrying the use of disposables rather than cloth (or vice versa), does not make for sensible policy. As such, the seemingly simple question of disposable versus cloth embodies in microcosm the inescapable difficulty of making policy about bigger concerns. There are always angles that we cannot foresee. Cleo Mussi, for instance, writes to the New Scientist:
.... I wonder whether the research comparing [cloth] to disposables took into account the fact that babies using cloth nappies tend to be toilet-trained day and night at a much earlier age – there is little more uncomfortable than a wet cloth nappy. A difference of six months to a year would lead to a child using 1100 to 2200 extra disposable diapers or nappies – a lot of extra landfill. Cleo Mussi, Letter to the editor of 'New Scientist', 12 December
Our environment and society are too complex and changing too rapidly for us to favour even one of two types of nappy. Our knowledge of existing and new scientific relationships is also rapidly expanding. Yet the way we make policy makes little allowance for such difficulties. Typically, a government (heavily influenced by corporate interests or ideological baggage) makes a top-down, one-size-fits-all decision, ostensibly based on fossilised science, and then moves onto something else, rarely revisiting or even monitoring (pdf) its performance.

When bigger challenges than nappies loom, this way of doing things generates commensurately bigger problems. Whether it's climate change or health or global conflict, neither government nor any single conventional organisation can know all about the relevant human and scientific relationships, let alone keep up with them. Nor can they anticipate the diverse effects their policies will have over both time and space. The complexities are too great, and any single body is going to be too pre-occupied with its image, the latest events, or its members' individual goals to care much about outcomes.

Yes, outcomes. Even obscure wrinkles, of the sort about which Mussi tells us, can have big, unforeseeable impacts. Only people who are continuously motivated to achieve our goals, to look at the effects of their initiatives, and to adjust their ideas accordingly, can develop the diverse, adaptive approaches that we need to solve our social and environmental problems. Social Policy Bonds are one way in which we could stimulate such initiatives. They have other benefits: most significant here is that issuers of the bonds do not need to specify how a problem is to be solved in order to get people started on solving it. Our goals are stable: the optimal ways of achieving them, especially when complicated by time lags, feedback loops, a multitude of known, unknown and unknowable variables, are not. We can and, in my view, should, issue Social Policy Bonds targeting such goals as dealing huge, urgent problems such as climate change or global conflict even though the ways in which they are to be solved are beyond - well beyond - the purview of our current policymakers and their paymasters.

15 December 2018

Health: begin with our goal

Jerry Muller writes:
If ...doctors are remunerated based on the procedures they perform, that creates an incentive for them to perform too many procedures that have high costs but produce low benefits. But pay doctors based on the number of patients they see, and they have an incentive to see as many patients as possible, and to skimp on procedures that are time-consuming but potentially useful. Compensate them based on successful patient outcomes, and they are more likely to cream, avoiding the most problematic patients. Jerry Z Muller, The Tyranny of Metrics, February 2018
It's perfectly understandable. People react rationally to the incentives on offer. For reasons that might be politically incorrect to suggest, we live in societies where the incentives are increasingly extrinsic - mainly financial - rather than intrinsic, which has to do with the satisfaction of doing a job well. It's a trend, rather than an absolute, but it leads to perverse incentives in many, perhaps most, professions including, as Mr Muller tells us, medicine.

What's to be done? How should healthcare professionals be paid? Leaving it to the market, as some might propose, is not a workable solution. One reason - there are others - is that the information asymmetry is too great. So: let's start with our goal: we want to see improvements in the long-term health of a country's population, and we want such improvements to be made as cost-effectively as possible.

My suggestion is that we apply the Social Policy Bond idea to health. This could be done at a national level. A government would issue Health Bonds that would become redeemable for a large sum once national health levels reach a targeted level, sustained for a period of years or decades. 'Health' would be measured by objective criteria, such as longevity, Quality Adjusted Life Years, infant mortality rates etc. Each of these measures would have to fall into a targeted range before the bonds could be redeemed. Investors would buy the bonds and in doing so find themselves members of a protean coalition all with the same goal: to bring about improvements in the country's health. Having done their bit and (hopefully) having seen the value of their bonds rise, they'd sell their bonds to people willing to bring about further health improvements.

Bondholders' goals would therefore be exactly congruent with those of society: to improve the nation's health with maximum efficiency. Note that, while bondholders might hold the bonds only for short periods, the bonds could be in issue, depending on their redemption terms, for decades. Unlike current policy, they'd be long term in nature.

There are other advantages, one of which is that the government (or other issuing body) would not dictate how our health goals shall be achieved, nor who shall do so. Doctors and hospitals would certainly have some involvement. But bondholders, being motivated to find the most efficient pathways, might find that, for instance, subsidising youth clubs in certain areas, or providing cheap taxi services in others, or giving out free e-cigarettes, or paying doctors differently, would be the least costly ways of seeing results. The possibilities are endless and no government, nor any conventional organisation, can be relied on to investigate, experiment and implement the innovative, diverse, adaptive long-term solutions that will help achieve our goal. Health Bonds would see the creation of a new type of organisation: one whose composition, structure and activities are entirely subordinate to society's goal: improving the nation's health as efficiently as possible.

03 December 2018

It's not just nappies

Nappies, cloth or disposable, just won't go away. I've blogged about nappies before (here and here). A recent article in New Scientist revisits the issue, citing an updated report done in 2008:

...if cloth nappies were washed in full loads, air-dried on a washing line and reused on a second child, they resulted in 40 per cent lower greenhouse gas emissions than using plastic disposable ones. These benefits are likely to be even greater today [ie 2018] now that half the UK's electricity comes from low-carbon sources. Are disposable nappies really so terrible for the environment?, Alice Klein, 'New Scientist', 21 November
The question posed by the article raises a general question here as to policymakers go about legislating or regulating in order to solve social and environmental problems. Most of these problems are complex: they have multiple causes whose significance changes radically over space and time. Our current policymaking system usually looks at a problem, or a symptom of a problem, tries to identify a cause, then goes about trying to legislate or regulate that cause so as to moderate its adverse effects. That works well when the relationship between cause and effect is easy to identify. But such is rarely the case today, as the nappies example shows.

It's not just nappies. Climate change is the same: we cannot rely on government to identify the causes of a complex problem, then do the right thing and, eventually, regulate it. There is too much scope for mis-steps along the way. The causes might be many and varied, with time lags and linkages impossible to verify. Legislating is nowadays a cumbersome and arcane process, and is often opposed or delayed by powerful interests that stand to lose if it goes forward. Results of policies are rarely monitored; still less are policies modified in the light of their impacts.

These flaws are inevitable in the way we make policy today. Fortunately, I believe there's an alternative. We need to focus on outcomes, and let the ways in which these outcomes be achieved be decided, on a continuous basis, by people rewarded only for achieving them. So: rather than government trying to deal with the problem of (say) landfill by commissioning a one-time study of the comparative benefits of cloth versus disposable nappies using fossilised data, it would instead target for reduction the volume of landfill. Rather than try to work out why there are more adverse climatic events, government should reward reductions in the number and severity of such events. And rather than try to work out some alleged 'root causes' of violent political conflict, we could instead reward the sustained absence of such conflict, whoever achieves it and however they do so.

Not only would this be intuitively more efficient than current the policymaking process, it would also be much quicker. It's taken decades to get to where we are now with regard to the causes of climate change and...adverse climatic impacts are worsening. And we haven't even begun to identify root causes of war. We don't even know if there are any....

What I'm advocating is, of course, the Social Policy Bond idea, whereby we issue bonds that become redeemable only when a targeted social or environmental goal has been achieved. Investors in the bonds would themselves work out the best ways of achieving these goals, and they would be motivated to do so efficiently and continuously. It's a simple idea, but the ramifications are many and varied, and I've written about some of them in this blog and on the Social Policy Bonds website. As I say: it's not just nappies. The traditional way of doing things just isn't working any more. It's time to focus on outcomes.