Spreadsheets of power
How economic modelling is used to circumvent democracy and shut down debate
Most people think it is hard to put a dollar value on a human life, but they’re wrong. It’s easy. Economists do it all the time.
Most people think that all human lives are equally valuable. And most think economic modelling is boring, irrelevant to their busy lives, and unrelated to how our democracy is functioning. They’re wrong about those, too.
About ten years ago, a lawyer rang to ask if I would do some (economic) modelling. “It depends,” I said. “What’s the job?”
“We want you to put a dollar value on the life of a dead mother,” said the lawyer. “We are suing a doctor for medical negligence, and the insurance company wants to value her life at zero because she wasn’t working. She had no future earning potential. Can you estimate the value of the housework she would have performed?”
I still feel sad when I think about it: for the family, for myself, and for a society in which asking such a question is not only acceptable but also necessary. The dilemma for the widower and the lawyer, and for me, was that if someone didn’t put a dollar value on the love and care that a mother gives her children, the father would wind up with even less money to care for the kids he would be bringing up by himself.
Of course, economists have no real way to value love and affection, so I valued ironing, laundry and child care instead. I got my hands on data about how mothers with three kids use their time. I found data on the price of buying individual household services like ironing, and the price of live-in maids and nannies. I forecast the age at which the kids would leave home. My forecast was based on a meaningless average of kids who do go to uni and kids who don’t. My spreadsheets were huge, complex, scrupulously referenced and entirely meaningless. Like all good forecasters I estimated the “value” of her life to the cent, and as happens in all good negotiations, the lawyers ultimately settled for a nice round number. The only good thing about the number was that it was bigger than zero.
The topsy-turvy “morality” of economics is built in to models that politicians, lawyers, economists and lobby groups use to persuade the public, in all parts of public life: models that say, for instance, that we can’t afford a price on carbon; that life-saving medicine for some people is “too expensive”; or that the loss of an entire species is justifiable if woodchip prices remain above $100 per tonne.
Everyone who uses economic models to excuse the inexcusable wants you to believe that the models are boring. The last thing they want you to do is to pay attention.
Many economists have calculated that it will be cheaper for the world to endure climate change than to prevent it. The models they use to draw this bizarre conclusion are built on thousands of assumptions about everything from the value of human life to the willingness of consumers to buy smaller cars if petrol becomes more expensive. If any one of those assumptions is wrong, the answer will be wrong. If hundreds of the assumptions are out, the answer becomes meaningless. (Some economists then argue that if hundreds of the assumptions are wrong then the errors might cancel each other out. Seriously.)
Imagine you were asked to model the costs of dangerous climate change. Imagine you were in possession of the likely number of people who will die as a result of storms, floods and droughts. Imagine you knew what countries they would die in, and how many years into the future. Would you value all of their lives equally? Would you assume that a Bangladeshi and an American life were “worth” the same? Would you think that the death of a child in 20 years’ time was worth as much as the death of a child in 50 years’ time?
In our democracy, these ethical questions are usually answered by economists, to two decimal places.
Most economic modellers do not assume that all human lives are equal. Bjorn Lomborg, for example, one of the world’s most famous climate sceptics, uses modelling that assumes the lives of people in developing countries are worth a lot less than the lives of Australians or Americans. While the US Declaration of Independence may declare that all men are created equal, most economic models assume that all men (and women) are worth a figure based on the GDP per capita of their country.
Late last year, Bjorn Lomborg asked to meet me, and I wondered whether talking to him would be good fun or a waste of time. It was neither: it was scary and illuminating. After 15 years as the smiling face of climate inactivists, Lomborg had raised his sights. His new mission was to ensure that governments also deliver inaction on global poverty alleviation, public health and gender inequality.
When we met, Lomborg proceeded to explain how his team of economists at the Copenhagen Consensus Center had decided that a number of the United Nations’ Millennium Development Goals weren’t worth pursuing. His tool of choice for defending such a position? Economic modelling.
Always be careful what you put in writing. Lawrence Summers is the president emeritus of Harvard University and was previously the chief economist at the World Bank. At the World Bank in 1991, he put his name to a notorious memo about the relative value of human life in rich and poor countries. Ironically, it begins with the words “just between you and me”.
The Summers memo spells out why it would be “welfare enhancing” (read: “good for humanity”) for all toxic waste to be dumped in poor countries:
The measurement of the costs of health impairing pollution depends on the foregone earnings from increased morbidity and mortality. From this point of view a given amount of health impairing pollution should be done in the country with the lowest cost, which will be the country with the lowest wages. I think the economic logic behind dumping a load of toxic waste in the lowest wage country is impeccable and we should face up to that.
Read: “Since toxic waste will inevitably make someone sick, we might as well make poor people sick, as doing so will minimise the amount of ‘foregone earnings’.”
The memo continues:
I’ve always thought that under-populated countries in Africa are vastly UNDER-polluted … The concern over [a pollutant] that causes a one in a million change in the odds of prostate cancer is obviously going to be much higher in a country where people survive to get prostate cancer than in a country where under 5 mortality is 200 per thousand.
Read: “If it takes years for some pollutants to make us sick, why not dump pollution in countries where people don’t live long enough to get sick from the pollution?”
Why wouldn’t you?
Summers claims that the memo was meant to be sarcastic and ironic, but, as anyone who has studied economics can assure you, the logic (if not the morality) is unquestionable. If you assume that humanity’s primary goal is to maximise world income, then killing poor people instead of rich people makes perfect sense.
You probably didn’t know economists had an assumption about humanity’s primary goal, did you? No wonder developing countries think that the developed countries don’t really care about their suffering as much as our inconvenience. We don’t.
Assumptions such as those made by Summers sit at the heart of the economic models that are regularly used to oppose carbon taxes, support free trade agreements and prevent the introduction of environmental regulations or more generous welfare safety nets.
Much of the power of economists is based on the public’s (understandable) lack of desire to read reports written in algebra. That’s why we like to use algebra.
While some cultures respect the elderly, economic models generally don’t. On the contrary, models such as those at the heart of Joe Hockey’s recent Intergenerational Report (IGR) are often used to create a sense of anxiety in the broader community about the “burden” that older citizens place on the “productive” members of the community.
Economic models are at their most powerful when only the powerful are aware of what they contain: thousands of assumptions that range from the immoral and implausible to the well-meaning but estimated. However they are made, the conclusions of a model are only as reliable as its assumptions. Garbage in, garbage out.
Back in 2003, the US public got a peek behind the modeller’s curtain, and it didn’t like what it saw.
The US Environmental Protection Agency (EPA) has generally valued all American lives equally when conducting so-called cost–benefit analyses of policy change. Generally. While pensioners are used to getting a discount on many things, American pensioners were surprised to discover that the EPA had experimented with discounting the value of their lives.
Given that air pollution generally kills the old and the sick, the consequences of devaluing the lives of the old was highly significant. By assuming that the old and the sick weren’t worth as much as an average American, it was harder to show that the benefits of tougher air-pollution regulations to the population were greater than the cost of tougher regulations to the car companies. Put simply, the EPA’s “pensioner discount” assumption could have saved air polluters a lot of money.
Not surprisingly, once the public realised that economists were devaluing their grandmothers’ lives, they got interested. Congress even passed a bill banning federal government funding for research that discounted the value of retirees’ lives.
In Australia, a human life is valued by the Department of Finance at $150,000 per year. That said, individual departments, consultants and lobbyists are free to use their own assumptions. Of course, citizens are also free to pore over the thick reports written by the modellers, looking for such assumptions, but few do. You might argue that it is the public, or its representatives in parliament, who should weigh up such “life and death” matters. But while the US Congress has expressed its dismay, the Australian parliament has never considered the issue.
Machiavelli told us that it is better to be feared than loved, because while love is transitory, fear is ongoing.
Politicians have always used fear to achieve their objectives. The “yellow peril”, “reds under the bed” and the “War on Terror” are all obvious examples, but have you ever thought about the objectives of the Intergenerational Report?
The government has used the IGR, the National Commission of Audit, the budget, the pre-election economic and fiscal outlook, and the mid-year economic and fiscal outlook to scare the public into accepting that we can never afford to tackle climate change or spend more on health, no matter how rich we become.
Most people, lefties in particular, have little time for economics. Modellers like it like that. Their findings and forecasts crowd out public debate because they, and those who pay them, have the power to limit the options we are allowed to consider.
Most people want to live long and healthy lives. They want their friends and family to do likewise, and they think that population ageing is a good thing for them and their parents. The IGR and its ilk, on the other hand, train us to believe that other people’s ageing is a burden.
The fear of the “costs of ageing” now lies at the heart of the campaign to limit our generosity to others and our ambitions for our community. Of course it would be great to provide world-class health care, we are told, but haven’t you heard that we are being crippled by the costs of ageing?
Joe Hockey recently said that his first (and most likely last) IGR would be so scary it would “knock voters off their chairs”. Look at the debt! Look at the deficit! The sky is falling!
The same government that couldn’t care less about climate change in 50 years’ time has conniptions about the deficit in 50 years. They cry “Won’t somebody think of the children?” with no sense of irony, proportion or consistency.
Unfortunately for Joe Hockey, his IGR is like a monster in a horror movie: the longer it is on screen, the less scary it is. In 2002, when Peter Costello first asked us to imagine how big the deficit would be in 2050, right-thinking citizens dutifully expressed concern. But, like most sequels, IGR episode IV was a pretty tired affair that generated more giggles than screams.
Australia is not only one of the richest countries in the world; Australians are among the richest people the world has ever known. While poverty certainly exists in Australia, the average Australian enjoys a material standard of living that few in the world can imagine. Still, we feel poor. Many Australians believe that retiring with less than $1 million in savings will leave them vulnerable in their retirement. That is exactly how the modelling behind the IGR is designed to make us feel: selfish, broke and vulnerable.
You would think it would be hard to make some of the richest people in the world feel poor, but with 13 years worth of dodgy modelling and powerful voices on side, it turned out to be pretty easy.
The finance industry spends a fortune telling us that you can’t “retire with dignity” on less than $50,000 per year (tax free), even if you own your own home. It’s certainly more scary than reminding baby boomers that they have a much higher standard of living than their kids. Just how people on the minimum wage of $38,000 manage to raise kids, pay tax, pay rent and maintain their dignity is the great unasked question of Australian politics.
According to the IGR, by 2055 our average incomes will rise by 75%, we will be living longer and healthier lives, and we will be paying a lot less tax. According to Treasury, by 2055 the average Australian will have an inflation-adjusted income of more than $100,000 per year.
All of those numbers come from the IGR, but they aren’t the numbers that Joe Hockey wants you to focus on. Economic modellers aren’t only expert in making dodgy assumptions; they also know how to put the numbers they want you to find up in lights, and how to bury the rest in the appendices.
My first job as an independent economic consultant was 20 years ago. I’d previously worked with other economists as part of a team but this was my first solo performance. I was a bit nervous.
After a brief phone call explaining what the client wanted, I spent days preparing for our first face-to-face meeting. When I had spent a few minutes outlining what I saw as the strengths and weaknesses of the possible methodological options, the client interrupted.
“Look, mate,” he said, “all I want is something about an inch thick. I want to walk into a meeting, slam it on the fucking table, and say, ‘According to my economic modelling …’”
The IGR isn’t an inch thick, but at 145 pages it is assumed to be thick enough to deter most people from reading it closely. (Not all the assumptions made by economists are unrealistic.)
Virtually everything you have heard about the IGR is wrong. It doesn’t say we are going broke. It doesn’t say we can’t afford to maintain the quality of our health system. And it doesn’t say that we have to cut the generosity of the age pension. That, however, is what people think it says.
The most important assumption made by Treasury is also the most surprising. The economists at Treasury have assumed that income-tax rates will be cut every year between 2020 and 2055 in order to ensure that we don’t collect too much revenue.
The IGR states that if we left the tax system alone for the next 40 years, the government would be flooded with a sea of revenue.
Treasury knows this. They admit it, safe in the knowledge that few people will read the enticingly titled ‘Appendix C: Methodology’, where it’s stated that the emphasis of the IGR “rested on pressures that demographic change was likely to impose on future government spending rather than the way these spending pressures may be financed (such as through increasing revenues or raising debt)”.
Read: “The purpose of the report is to get you worried about rising costs. If the report pointed out that budget revenue was likely to rise faster than budget outlays, people wouldn’t be scared.”
Through careful editing, strategic summarising and a few tricks of data presentation, Treasury is putting out the message that all Australians should be worried – about how future people who will be far richer than us, and pay lower taxes, will afford to pay for their health care.
Well played, Treasury. Well played.
As they put it in Appendix B: “Projections in this report have been developed using a range of assumptions … There are significant uncertainties around these assumptions. As such, this report should not be viewed as a forecast.”
But we can’t just keep spending and spending forever, I hear you say. Surely there are clouds on the horizon for which we should be planning.
Australia is not only one of the world’s richest countries, but also one of the lowest taxed developed countries. If we wanted to have some of the world’s best health, education and public transport systems, we could. If we can’t afford world-class services, who can?
But Peter Costello, Joe Hockey and the modellers at Treasury have never wanted to invest in world-class services. They think we should cut taxes, which is fine: in a democracy, people should be free to argue for whatever levels of taxing and spending they want. But the strategy of using modelling to scare us about future budgets effectively removes a range of choices from our democratic menu. It’s OK that Costello and other conservatives want to live in a low-tax country that discourages sick people from going to the doctor, but it’s not OK to use public servants to trick the public into thinking that this is not a choice. In our democracy these debates aren’t won with persuasion, they’re won with dodgy modelling designed to convince Australians that we couldn’t possibly afford to spend as much on health as Austria or Canada does.
The IGR doesn’t show that we are broke. It shows that if we want to keep cutting taxes we will have to cut spending.
Democratic decision-making is pretty ordinary at the best of times, but it is downright awful when key choices are concealed from the public and key evidence in support of a government’s preferred option is simply fabricated. Fabricating such evidence is increasingly the role of economic modellers.
A few years ago, I was contacted by representatives of the Bulga Milbrodale Progress Association. Bulga is a nice little town near a big coalmine. Its 400-odd residents were in a fight against Rio Tinto, one of the world’s largest mining companies. And, like many people embroiled in a big fight with a powerful company, they realised they were really in a fight about economic modelling.
When they asked me if I was willing to look at an economic impact statement prepared on behalf of the mining company, I responded cautiously. I was keen to help, but told them I would only get involved if they were willing to directly attack the assumptions underpinning the modelling used to estimate the benefits of the mine. The environmental movement had spent decades avoiding a direct attack on the claimed economic benefits of mining, preferring instead to try to counterpose a value on the possums, frogs and trees that are inevitably harmed. I like possums, frogs and trees, but I think attempts to value them are as arbitrary as attempts to value human lives.
They agreed to tackle the miner’s claims about economic benefits head-on, and they won the first ever court case to stop a mine expansion or approval in New South Wales. (They won again when Rio Tinto appealed to the Supreme Court, but lost after the state government changed the law to make it easier to approve mines.)
It’s never a good look to bulldoze a tiny historic village in the name of foreign shareholders. Rio Tinto knew that it needed to distract attention away from the profits that it would make destroying the village of Bulga and towards the “benefits” that would flow to the community that was adamantly opposed to the project.
That’s where the modellers came in.
Rio Tinto’s economic consultants wrote a fat report nobody was supposed to read. It made the absurd claim, in its executive summary, that if the mine extension were approved it would create 44,000 jobs: a figure that is more than double the entire NSW mining industry’s employment level today.
The 44,000 figure is obviously a deception. The authors had redefined a job as a “person year” (i.e. a job for one person for one year) so that they could multiply their real job-creation estimate by the number of years in the life of the mine to get a bigger number. It’s a simple trick, and it only works when people are too busy to read a thick report. The thicker, the better. But that’s not all they did.
You have probably heard that mining creates a lot of “indirect jobs” in construction, manufacturing and retail. You may not have heard that building hospitals and employing nurses would have exactly the same indirect benefits. Mining isn’t the only industry that can create “indirect jobs”, it’s just the only industry that spends millions of dollars each year paying economists to estimate such “benefits”, and millions more spruiking them to the media.
In most cases, economic modelling isn’t used to help us understand the costs and benefits of major projects or policy changes. It is used to conceal them. Rather than outline and discuss the wide range of winners and losers, debate the pros and cons, and make transparent decisions, economic modelling allows all of the messy bits to be hidden.
In 2011, Denmark’s general election saw its centre-right government tossed out of power, to be replaced by a minority centre-left coalition led by the country’s first female prime minister, Helle Thorning-Schmidt.
Bjorn Lomborg’s Copenhagen Consensus Center was one of the first casualties of the change of government. When it was announced that its more than $1 million in funding would be cut, Lomborg visited the new prime minister, urging her to reconsider the government’s decision. “I’d love to show you how the Copenhagen Consensus is a good idea,” he was reported as telling her.
“I think that probably might be right, Bjorn,” she reportedly responded to the sceptical environmentalist. “But I will just get so much more mileage out of criticising you.”
Costs and benefits can be calculated any number of ways, and the modeller’s assumptions are crucial to the end result. Lomborg had confidently assumed that the Danish taxpayer would continue to fund his work. His cost–benefit analyses had found that more effort should be put into free trade and less money spent on tackling poverty and climate change. But, as with all such efforts, garbage in, garbage out.
There is a role for economists, and economic modelling, in public debate. Its role should not be to limit the menu of democratic choices. Instead it should be to help explain the trade-offs.
Good modellers aren’t afraid of explaining their assumptions. The clients who pay best, however, don’t want the best modellers. They want people who can write a fat report to slam on the fucking table.