Risk communication

Discussing asymptomatic women aged 40-50 who have participated in mammography, Gerd Gigerenzer writes in his excellent book Calculated Risks (pp41-3):

The probability that one of these women has breast cancer is 0.8 percent. If a woman has breast cancer, the probability is 90 percent that she will have a positive mammogram. If a woman does not have breast cancer, the probability is 7 percent tht she will still have a postive mammogram. Imagine a woman who has a positive mammogram. What is the probability that she actually has breast cancer?

Of 24 physicians who were given this information only two gave the correct answer - which is 9 percent. Probabilities are difficult for most of us to get a handle on. Gigerenzer is a persuasive advocate for instead using natural frequencies:

Eight out of every 1000 women have breast cancer. Of thse 8 women with breast cancer, 7 will have a positive mammogram. Of the remaining 992 women who don't have breast cancer, some 70 will still have a positive mammogram. Imagine a sample of women who have positive mammograms in screening. How many of these women actually have breast cancer.

It's much easier to see from this that only 7 of the 77 women who test positive actually have breast cancer, and indeed most of 24 (different) physicians given the information in this form, estimated correctly. Gigerenzer also discusses AIDS testing and the often tragic reaction of people who test positive when they, their testers and their counsellors know nothing about how frequently such positives are false:

Since the first AIDS cases were described in 1981, more manpower and money have been poured into researching HIV than any other disease in history. Little, in contrast, has been done to educate the general public about what an HIV test result means. (p139-40)

Ignorance about risk bedevils the law courts as well:

Many students who spent much of their life avoiding statistics and psychology become lawyers. Out of some 175 accredited law schools in the United States, only one requires a course in basic statistics or research methods.... [S]tudents who excelled in critical thinking could not evaluate whether a conclusion drawn from statistical evidence was correct or incorrect. (p159)

I think the lessons from this are important for policymaking, and not just because many of our policymakers used to be lawyers. One lesson is that it's quite possible for even well-meaning professionals to have no idea about statistics and risk. As our economies and societies grow ever bigger and more complex, policymakers will rely more and more on statistics. Inferences drawn from them need to be robust. My feeling is that the ignorance that Gigerenzer documents thrives in the compartmentalized, specialized policy environment we have today. Doctors, lawyers and politicians don't know very much about risk and neither does most of the public. There's very little monitoring of policies for effectiveness and there's very little incentive to get policies right. You can't (easily) legislate for effective risk communication and understanding, but what you can do is throw the achievement of social goals open to the market - by using Social Policy Bonds for example - so that errors of the sort that Gigerenzer documents do not persist and entrench themselves.

Marjane Satrapi

Graphic novelist Marjane Strapi was interviewed in April 2005 for Salon Magazine (gated, but the full text is here). She says:

If I have one message to give to the…American people, it's that the world is not divided into countries. The world is not divided between East and West. You are American, I am Iranian. We don't know each other, but we talk together and we understand each other perfectly. The difference between you and your government is much bigger than the difference between you and me. And the difference between me and my government is much bigger than the difference between me and you. And our governments are very much the same.

This is a compelling insight. It's especially poignant ("We, the people....") that Ms Satrapi can accurately point to the gap between the American government and American citizens. Big government need not necessarily be a problem in itself, but it tends to come with remote government - which, I believe, is. What's more, big government is self-entrenching. It is comfortable dealing with (and accepting campaign funding from) big corporations. Big business of all sorts enjoys explicit subsidies, as well as a favourable regulatory environment, and the implicit subsidies of a government-funded infrastructure and (often) economic protection. Governments confuse the fortunes of big business with those of the wider economy, and those of the wider economy with those of society. One result is that the individual citizen in most democratic western countries feels as remote from decision making as does the average Iranian citizen.

Incentives to engage

According to Jim Giles, writing in the New Scientist (subscription), New York this year became the first city in a rich country to try to alleviate poverty by offering cash incentives to improve people's engagement in areas such as education, health and employment. Mexico was the pioneer. Top-down projects, such as subsidies for staple foods and healthcare were mostly unsuccessful. So the government gave cash payments to low-income families to be spent however they wanted, provdied they behaved in approved ways. FOr example, a family could earn about $20 a month by enrolling a child in primary school and ensuring that s/he attended regularly. Similar payments were made if children had regular health check-ups.

In the rich countries it is mainly the US that uses such incentives, and there only in a few isolated drug-treatment programmes. Whether they succeed in stopping drug abuse in the long term is uncertain.

It's a controversial approach, but one of which I approve.I'd prefer governments not to make the payments directly, but rather to set broad health, education and employment targets, and let the private sector work on achieving them. But accepting that reducing substance abuse (for instance) is an end in itself, I'd encourage the disconcertingly rare approach of setting a target and doing whatever is necessary to achieve it. The chosen methods may seem controversial, counterintuitive or as in this case, a subsidy to the undeserving or the dissolute, but so long as they are ethical and legal I think it would be irresponsible to rule them out.

Slow news day

My blog is worth $2,258.16.
How much is your blog worth?

Ludicrous

The Accident and Emergency (A&E) departments in English hospitals have to ensure that 98 percent of patients transferred or discharged within four hours. If they fail, they are subject to financial penalties. One result is that, since these targets were imposed, more patients are transferred to hospital wards 'just in case'. This is costly in resource terms, but it benefits hospitals who receive as much as £1000 per admission, compared with about £100 for a patient treated in A&E. And, as The Times puts it:

The increase in admission through A&E could have another explanation, apart from the four-hour target. To admit more patients is greatly in the financial interests of hospitals because under payment by results they get paid much more. Using the system in this way is called “gaming” within the NHS and is frowned upon.

This the sort of nonsense that happens when a government imposes Mickey Mouse micro-targets. The letter of the intent is strictly adhered to; the spirit is ignored. I do agree that we need some sort of target for government funds, but such targets must be broad and meaningful to ordinary people. A&E throughput rates are irrelevant to the health of the population; if government is concerned about health then it should target health. Of course, defining health and measuring it is more difficult than measuring the length of stays in A&E departments, but with sufficient ingenuity - of the sort that is currently applied to gaming the system - it can and should be done.

Politics as an Enterprise System

In a review of The Social Life of Information by John Seely Brown and Paul Duguid, Simon Head describes:

"enterprise systems," or ES, which bring together computer hardware and software to standardize and then monitor the entire range of tasks being done by a company's workforce.

It is ES that Wal-Mart has applied to the retail economy, to the great benefit of its shareholders.

Among manufacturers, wholesalers, and retailers like Wal-Mart, ES offers obvious economic advantages. It relies on electronic tags, sensors, and "smart" chips to identify goods and components at different stages of the production and distribution chain, a practice that has brought enormous gains in productivity. Such innovations allow managers to find out immediately not only that production and distribution are falling behind schedule, but also why.

But increasingly ES is being applied to bureaucracies, white-collar business, and universities. ES technologies reduce complex human activities to a series of processes that can be mapped out and programmed by a computer. Writing about the US, Mr Head continues:

Nowhere have these technologies been more rigorously applied to the white-collar workplace than in the health care industry. The practices of managed care organizations (MCOs) have provided a chilling demonstration of how enterprise systems can affect the work of even the most skilled professionals, in this case the physician. The goal is to standardize and speed up medical care so that insurance companies can benefit from the efficiencies of mass production: faster treatment of patients at reduced cost, with increased profits earned on increased market share.

This seems to work only from the very narrow perspective of the MCOs' accounts. Patients experience similar frustrations (and worse) to those that all of us feel when ringing a call-centre - where ES is also widely used.

And what about the workers? Mr Head mentions also The Culture of the New Capitalism, in which Richard Sennett

describes how the widespread use of enterprise systems has given top managers much greater latitude to direct and control corporate workforces, while at the same time making the jobs of everyday workers and professionals more rigid and bleak. The call centers of the "customer service" industry, where up to six million Americans work, provide an egregious example of how these workplace rigidities can make life miserable for employees. At call center companies such as AmTech and TeleTech, call center companies to whom many corporations outsource their "customer relations management," agents must follow a script displayed on their computer screens, spelling out the exact conversation, word for word, they must follow in their dealings with customers. Monitoring devices track every facet of their work: minutes spent per call, minutes spent between calls, minutes spent going to the bathroom. ...

The most powerful passages in Sennett's book describe how these unnerving changes are destroying aspects of white-collar employment that he believes are essential to the well-being of workers, whether they are nurses, call center agents, bank officers, or mid-level managers at Con Edison. He describes how the spread of ES has resulted in a declining emphasis on creativity and ingenuity of workers, and the destruction of a sense of community in the workplace by the ceaseless reengineering of the way businesses operate. The concept of a career has become increasingly meaningless in a setting in which employees have neither skills of which they might be proud nor an audience of independently minded fellow workers that might recognize their value.

This is bad enough, but what happens when bureaucracy and politicians adapt ES for their own purposes? In New Zealand and Australia it appears that the independence of public servants and their ability to offer impartial and objective advice to politicians is diminishing. Policymaking becomes a top-down, hierarchical process, designed to maximise the popularity - as measured, with ever less credibility by general elections - of the ruling political party. Officials become subject to the ennui that Sennett describes. Flair and originality are discouraged, process is king. We are all Wal-Mart employees now!

All this serves to widen the already big gap between politicians and ordinary people. Slipping through the cracks are such vital but difficult-to-measure concerns as the wellbeing of society. A Social Policy Bond regime could close this gap. The key is to target outcomes that are meaningful to ordinary people. Such outcomes would be inextricably linked to people's wellbeing. Goals would originate from the public, who would be motivated to participate in politics because their views would count. Such an approach requires that politicians relinquish some of their power. Frankly, and mainly for that reason, it will be a while before it happens. But the alternative of an ever-widening gap between politicians and the people they represent and the continuing alienation of people in government service at any level from meaningful employment is too dreadful to contemplate.

Outcomes versus emotions as policy drivers

What should drive policy? Not ideology, I don't think, because it essentially faith based; the faith being that approaches that succeeded in the past will succeed today. (Paul Krugman has a column today on how conservative ideology has done much to precipitate the mortgage disaster in the US.) Not spending, because pumping money into activities or institutions is no guarantee efficiency in delivering outcomes will be maintained or improved. I'm also suspicious of top-down restructuring having worked in an organization that was restructured eight times in my 17-year career there. The immediate result was a loss of morale, losses of well-qualified employees, and waste of resources. There were short-term accountancy-type gains, but the main discernible goal seemed to be to enhance the curriculum vitae of those who planned the restructures. It's not only me who is skeptical of these sorts of policy driver. But there are worse, and one such is emotion.

A recent article on Science Daily is titled "The Effect of 'In-Your-Face' Political Television on Democracy". It reports on research showing that:

[T]he incivility and close-up camera angles that characterize much of today’s “in your face” televised political debate also causes audiences to react more emotionally and think of opposing views as less legitimate.

I think this finding is important. Television is vastly influential in politics. TV corporations have their own imperatives, and these have everything to do with audience figures (and subscription and advertizing revenue) and very little to do with fostering the mutual respect of opposing sides in political debates. Arguments are polarized, attitudes become extreme on all sides. In the US the influence of emotion as a policy driver seems to me to show in the attitudes towards, amongst others,smoking, guns, abortion and capital punishment. To this outsider there seems very little mutual respect on either side of those arguments, and the disease seems to be spreading within the US to immigration and welfare. And outside the US? I don't really know what television is like in other countries except perhaps New Zealand, where there also seems to be a descent into incivility.

It seems that emotion is supplanting other policy drivers. It's not too much of a surprise, because society is growing ever more complex as are the relationships between cause and effect in social and environmental policy. Emotion is easier to communicate and to exploit. But as a policy driver it has obvious faults. It's far too easy to manipulate for mercenary and more sinister ends. I'd much prefer to see meaningful outcomes drive policy. These could bypass the complexities of our economy and society, so they would be easy to understand. It's far simpler, say, to target violent crime, or climate change, than it is to make a case for (say) subsidizing leisure centres for youths or urging poor countries to stop building coal-fired power stations. These actions might be necessary and efficient, but it should be up to the market to make the case for carrying them out, rather than remote, cumbersome and monolithic central government. But that's not the only advantage of targeting outcomes: the other is that it would be direct and accessible to ordinary people. It would appeal to people's rationality, make us aware of trade-offs, encourage public participation and buy-in, and engender mutual respect amongst people with different views - in stark contrast to appealing to people's emotions.

Ideology and AIDS in Africa

The history of the response to African AIDS can be divided into two phases: (1) fiddling while Rome burns, and then (2) trying to use the fiddles to put out the fire.

So writes William Easterly in the New York Review of Books. The World Bank and aid organizations were slow to act.

Western scientists flew into Africa, collected blood samples, and flew out, seemingly much more interested in getting recognition in the Western press than in communicating useful and sensitive knowledge to African leaders and the public.

Western scientists were tactless; western media played up to this:

[B]eing accused of promiscuity and having Africans labeled as the equivalents of Typhoid Mary did not make their leaders or the general population all that receptive to messages from Western scientists on how to confront the epidemic. Many Africans reacted with a mixture of denial and conspiracy theories. Maybe the CIA had targeted Africans during the cold war with a scientifically engineered virus....

Helen Epstein in The Invisible Cure, the book that Mr Easterly is reviewing, understands that the AIDS crisis in Africa is caused by 'concurrent relationships'. As Easterly puts it:

To oversimplify a little, Africa's AIDS tragedy is that it combines greater Western-style sexual equality for women with social norms that permit simultaneous long-term sexual relationships for both partners.

Infection rates in Uganda fell, largely because of its crucial 'Be Faithful' message. Tragically, this message has been ignored by ideologues of both the left (who favoured condom use) and the right (who favoured abstinence). A huge share of the current western effort:

has been concentrated on getting antiretrovirals [ARVs] to those in Africa with full-blown AIDS. There is nothing wrong with the urge to treat the sick, but in practice it has crowded out nearly every other response to the epidemic. ARVs are now reaching only a tiny minority of those in need and it will never be feasible to treat everyone. .... The "Be Faithful" message was neglected because it was not of interest to the bureaucracy concerned with AIDS. As Epstein muses acidly: "Zero Grazing" had "no multimillion-dollar bureaucracy to support it."

This appears to be another distressing case of well-meaning, hard-working people being hampered by their own ideology in their genuine efforts to alleviate a human disaster. I think this is a clear case where a Social Policy Bond approach that rewarded people for reducing AIDS infection rates, however they do so, would be far superior. Aid organizations, and others, would direct funds impartially to where they would generate the greatest return per dollar outlay. Idelogues would go somewhere else: Social Policy Bonds would enlarge and motivate the pool of people who are solely concerned to reduce the incidence of AIDS, rather than validating their worldview.

Targeting longevity

If ever Social Policy Bonds were to be issued, they may well make use of existing sources of data that can be used, with some massaging. Though climate stability doesn't seem to be explicitly targeted, for example, there are large and increasing volumes of data on the world's climate and its variations over time and space. Now I read that Goldman Sachs is to compile publish a mortality index, tracking a "monthly a pool of 46,290 anonymous U.S. citizens over age 65". The rationale has nothing, unfortunately, to do with Social Policy Bonds, but this is the sort of information that would be helpful in targeting longevity, or some refinement of it, perhaps in countries where it is a reliable indicator of wellbeing. Note that, if the financial benefit that could be accrued by manipulating mortaility rates downwards were sufficiently high, the index could indeed stimulate longevity-reducing initiatives.

Climate change and discount rates

A not-too-technical discussion on the discount rate to be applied to the dmaage caused by climate change is here. As I say in my comment there, I can't help thinking that the cost-benefit method, in being applied to climate change has little relevance or meaning. For one thing, it's being applied to the so-called costs of mitigation, but the uncertainties extend to the relationship between gas emissions and climate change. I'd prefer to see climate change tackled directly and in fact targeting climate stability, rather than gas emissions, and subordinating policy to that would make life much easier. I suggest agreeing on a target for climate stability, and contracting out the achievement of a stable climate to the market. This means no prejudging of how best to achieve the goal: mitigation might be necessary, but there will other, possibly more efficient solutions, as yet unthought of by those who would commit us to gas reductions and only gas reductions.