16 August 2008

Irreversible mistakes

This past week, NYT has been doing an exposé on the plight of sick illegal immigrants in the US. The consequences of government mismanagement »in this case was so horrible that I could not bear to finish reading it.

Time and again, it seems to me that policy makers and implementers around the world do not understand enough that (1) they are dealing with people's lives; and (2) their split-second decision ("You're faking ill; I'm not sending for the doctor") can have irreversible consequences on people's lives.

I attribute this to the fact that no time is spent on discussing the art and ethics of decision-making when one first becomes a public servant.* Add on the ohh-so-fashionable imposition of KPIs, and you get people averse to erring on the side of humanity**. Alarmingly, I am not even sure if law enforcement and social services officers are given a rule of thumb such as, "When you are unsure and 'no' has severe, irreversible consequences, say 'yes (i'm sending for the doctor)'."

For a broader discussion on the difficult ethical issues surrounding healthcare for illegal immigrants, see »the story of Mr Jimenéz.




* Certainly not in my case.
** See also »this article. I used to say 'on the side of caution' »here, but 'caution' is too broad. The point is to prevent lethal mistakes.

4 comments:

  1. Jo, have you read this article about valuation of human life by govts so they can calculate the costs versus the life-saving benefits when drawing up particular regulations?

    Reminded me of Dr. X's saying: "Once you put a price on everything, you know the value of nothing."

    -----------------------------------
    What is human life worth?
    By Euston Quah & Chia Wai Mun , For The Straits Times (12 August 2008)

    AN AMERICAN'S life has become cheaper. How much cheaper? By about US$1 million (S$1.4 million), according to a US environmental body.
    A public outcry erupted when the Associated Press (AP) reported on July 10 that the US Environmental Protection Agency (Usepa) had lowered the value of an American's statistical life from US$7.8 million five years ago to US$6.9 million now.

    Not surprisingly, the Usepa did not make the devaluation public. But AP's Seth Borenstein discovered the devaluation after he reviewed US government cost-benefit studies over the past decade. Lowering the value of a life will affect government policies profoundly because the less a life is worth to a state, the less need it will feel for regulations to protect life.

    The issue of valuing lives also cropped up in Singapore during the recent debate on organ trading. Which matters more: saving people or sticking to an ethical code? Which should we choose: more dialysis machines or more kidney transplants? Answers to such difficult question come down in the end to the worth we assign lives

    The thought of putting a dollar value to a human life may provoke moral outrage but the process is necessary for good public policy. No country has an infinite amount of money and resources to spend on protecting and extending each citizen's life. At some point, choices have to be made in such areas as health care and safety regulation.

    Policymakers out to get the biggest bang for their taxpayers' buck must decide how much resources they will allocate to prevent unnecessary deaths rather than, say, improve education or public housing. The value of a statistical life reflects what people are willing to spend to reduce small risks of death. It is a measure used widely to evaluate public policies in medicine, environmental regulation and transportation safety.

    Everyone accepts some measure of risks in his or her life. Some of these risks can be avoided by spending money. When a person tries to avoid potentially fatal risks, or accepts compensation to take such risks, he implicitly defines a trade-off between wealth and a lower or higher chance of death.

    There are many methods of valuing a life, but most centre on one idea: The value of a statistical life should roughly correspond to the value that people place on their lives in their private decisions.

    Suppose workers face a one in 10,000 risk of being killed each year and that they accept this risk in return for an extra $200 in annual wages. The statistical value of life then becomes $2 million (200 x 10,000).

    But this number does not imply that people would accept death if paid $2 million or that they would come up with $2 million to prevent a certain death. Rather, it captures the amount which would make people consider a small change in the risk of death.

    Government agencies put a value on human life so they can calculate the costs versus the life-saving benefits when drawing up particular regulations. If they set that value too low, regulations to protect life - such as stringent airline safety and tighter pollution restrictions - would start to look like more trouble than they are worth.

    The value of a statistical life tends to correspond with per capita income. With a per capita GDP of US$45,845, the US has a statistical life valued at US$3.6 million (the average of 39 studies conducted in the US on the value of a statistical life). Australia, with a per capita GDP of US$36,260, has a statistical life value of only US$2.2 million. Among newly industrialised countries, Taiwan and South Korea have the lowest - US$1 million and US$646,000, respectively. What is Singapore's?

    In December last year, we conducted a study to determine the value of a statistical life here. We asked people how much they were willing to pay for a small fall in their chances of dying.

    From a sample size of 800 respondents, we conducted personal interviews with residents in seven housing areas - Yishun, Redhill, Tampines, Boon Lay, Bukit Timah, Choa Chu Kang and Sengkang. We estimated the value of a statistical life here to be between S$850,000 (US$606,00) and S$2.05 million.

    Being a relatively advanced economy, with per capita purchasing power parity GDP of US$49,714, the value was close to that of South Korea and Taiwan but was surprisingly lower than America's or even Australia's. However, while the figure is lower here than in other advanced economies, the sum of S$850,000 to S$2.05 million is not small when translated into public project evaluations.

    Of course, other economists may disagree with the figure we arrived at. But whatever the disagreements, we should try to put a value on life here for it would help us refine public policy in a wide variety of areas.

    After the Sept 11 tragedy, the US government issued guidelines for compensating the victims' families, dividing the payouts into economic and non-economic parts. It must have been extraordinarily difficult to stick a monetary figure on the intrinsic value of life in such circumstances. But the US government had to do it because the victims' families demanded some form of compensation for their loss and the public expected it.

    Ultimately, the question is not whether we should take on the challenge of ascribing value to life but how we should do it.

    ReplyDelete
  2. Jo : I wonder which country does these sorta things well?

    ReplyDelete
  3. @ Cindy: Thanks for the post! Yes, I met someone who was actually part of the value of statistical life project for S'pore. Interesting idea, but still, the survey asks hypothetical scenarios and I think we should just draw general trends from the results and not read too much into the S$850,000 and S$2m figures. In any case, the value of statistical life addresses the case of Mr Jimenez in Florida, but not that of Mr Ng in New York. Truth be told, Mr Ng suffered too much unnecessary pain as a result of presumption by the detention center staff/system; we didn't get into the stage of discussing who should beear his medical costs because he died in pain before that. To me, that unnecessary and severe suffering is the most untolerable and frustrating part...

    ReplyDelete
  4. @ Kelvin: I'm not sure at all. Even though this happened in the US, I'm aware the problem is not exclusive to it. It is also not coincidental that such issues are first brought to light in the US context because US still retains a fair bit of investigative journalism based on my impression.

    But well, if none of us are doing it well now, let's start to do so. Some imperfections and failures happen because it was a conscious tradeoff; some because of presumption and negligence.

    ReplyDelete