Talking Pints: Iraq and the Law of (Misleading) Averages

An oft heard remark about Iraq today (at least where I hang out) is something along the lines of “Well, it may be bad over there, but at least they (the Iraqi people) are better off than they were under Saddam.” Such a response strikes me as simultaneously reasonable (it may be true) and false, insofar as it may be little more than the ‘last line of defense’ justification of many folks for what is increasingly seen as a losing proposition. Bush’s recent declaration that finishing the war will be effectively ‘someone else’s problem,’ seems only to strengthen the latter interpretation. But let’s take the claim of “at least they are better off than they were under Saddam” seriously for a moment. For if it is true, then one might hope that the future is not so bleak after all.

There seem to be (at least) two issues tied up in the statement that “they are better off than they were under Saddam.” First, that the ‘quality of life’ of the Iraqi people is, on average, better, with standard indicators such as per capita GDP, and the number of people receiving basic services such as electricity etc. moving in the right direction since the invasion. Second, that regardless of the quality of life, one’s actual life is better preserved today, despite the violence that seems ever present, than under the old regime. While seemingly appealing, the problem with both claims, I suggest, is that they tend to rest, at least implicitly, on calculations of ‘averages’. Unfortunately, focusing on such indicators and sampling for averages to make meaningful comparisons may hide more than it illuminates.

Take the first claim, that quality of life indicators are (on average) moving in the right direction. If one examines the available statistics then the picture presented seems to back up this claim. Regarding GDP growth, the US Department of State notes that a year before the invasion “Iraq’s per person income had dropped from $3,836 in 1980 (higher than Spain at the time) to $715 [in 2002] (lower than Angola),” which is pretty poor by any standard. In contrast, in 2005 the State Department reports that “Iraq’s GDP is projected at $29.3 billion…up from $18.4 billion in 2002.” Moreover, “the IMF projects Iraq’s economy to grow by 10.4% in 2006.” Regarding electrical power as another indicator of progress, the same Department of State report notes that “more than 2,000 megawatts (MW) of generation capacity have been added or rehabilitated. One hundred fifty planned and ongoing projects worth $800 million will add more than 600 MW of additional generation capacity and improve the distribution of power to more than 2.1 million people.”

These are indeed successes, but does it mean that Iraqi’s are better off, on average, since the invasion? If it is at least plausible that the sanctions placed on Iraq by the UN from 1991-1999 lowered GDP by as much as 75 percent, or some equally large amount, then the recovery to the current level of per capita GDP of $3,400 seems somewhat less impressive. Moreover, confusion abounds as to what the real figures for Iraqi GDP actually are. For example, the CIA estimates the per capita GDP of Iraq in 2001 at $2,500.00 and in 2003 at $1,600.00, which makes the 2002 figure of $715.00 used by the State Department seem rather deflated. Regardless, similar to the ‘miracle of Reaganomics’, if you throw yourself out of a building and break both your legs (in 1980-82), the ability to crawl away on your elbows (in 1984) could be considered a success -– on average.

Regarding electricity supply, the recent growth in Iraqi generation capacity has to be seen against, not just recurrent insurgent attacks, but against the decrepit state of Iraqi infrastructure at the time of the invasion (due in part to sanctions), and the orgy of looting that eviscerated what was left of that infrastructure in the first six weeks after the invasion. Seen in this light, what restored power there is may be far less than is needed, with some estimates arguing for 6 gigawatts of new capacity to meet current demand in the context of a projected current shortfall of 1.1 Gigawatts despite the new capacity noted above. On average then, electricity supply may be higher today than it has been since the mid 1990s – on average - but that’s not saying much.

Third, what is making Iraq better off are not its oil revenues, which are up but wholly insufficient to rebuild, or the export of dates (the export success of the past few years apparently), but massive foreign (US) aid. Given the falling popularity of the war in the US and the long run costs of the war being projected as being as high as two trillion dollars it is unlikely that this ‘development by aid’ can be supported in the long term. Given all this, even if the optimistic statistics and projections are correct, which they are unlikely to be (given the bogusness of most forecasting), then it is hard to unambiguously make the case that the Iraqi’s are materially better off, on average, than they were under Saddam. Specifically, since the current recovery is contingent upon unlimited foreign largess that can disappear rather quickly, thus skewing the average quite drastically, it is not clear that the average a year or two from now will be anything like the average today.

What then for the other claim, that physical security is better now than under Saddam? Here the picture is equally complex. If we are to compare the threat to individual life today to that during the old regime, we must remember when Saddam et al., committed the majority of these murders rather than average over the life of the regime. Doing so would be to average out, for example, deaths in the Soviet Union over 70 years, and thus equally blame Stalin and Gorbachev. The problem is that ‘average’ deaths ‘now’ versus ‘then’ are a problematic indicator for comparison, (even if, unlike the USSR, Saddam was in power for the whole period).

Consider that the major ‘killing periods’ of Saddam’s regime were the ‘Anfal’ campaigns against the Kurds in the late 1980s where it is estimated 180,000 were killed, and the 1991 revolt where some 60,000 were killed. Add in the 500,000 Iraqis slaughtered during the Iran-Iraq war, plus the estimated 50,000 or so people murdered at other points, and you end up with about 790,000 people killed. [Photo shows Kurdish victims of the poison gas attack at Halabja.]

Compare this to the numbers given in The Lancet study of 2004 (or the UN study of 2005) where it was estimated that 100,000 had died as a result of the invasion (slate), or the (more documented and less estimated) study of Iraq Body Count of between 33,000 and 38,000 deaths since the invasion, and it seems quite simple to conclude that one’s safety today is greater than it was under the old regime. Indeed, one estimate places the old regime death rate at “between 70 and 125 civilian deaths per day for every one of Saddam's 8,000-odd days in power.” (In contrast, see the Bode Miller problem I discussed last month). As such, things may, on average, be better today than the were in the 1980s or 1990s, but we should not expect anyone unfortunate enough to be living Iraq today to calculate their position relative to the average and be thankful for it.

Specifically, the data in Iraq, especially regarding political murders, is especially lumpy. Rather than their being an average number of murders per day by Saddam that people could expect, there were brief periods of intense violence punctuated by long periods of relative inactivity. In contrast, what we seem to have in Iraq since the occupation is constant and increasing levels of violence, even if the average rate is lower. Which situation then is harder to deal with? Reflecting on my wife’s family’s experience made me think about this question.

My wife was born and raised in East Germany, Stasi and all. Indeed, her uncle spent a few years in prison for going against the regime. However, the rest of her family did not. The reason was simple. Dictatorship or not, the East German regime had certain rules and norms that were obvious to all its citizens. If you obeyed these, you had a Stasi file like everyone else, but probabilistically, if you did not break the rules or violate the norms, you were left alone. There was little about the lived environment that was radically uncertain. Saddam’s regime may have been far more vicious than the former East German communists, but given the lumpiness of the data on who was killed and when, it may have been a reasonable assumption that if you were neither in the army in the 1980s, not a Kurd or a Southerner in revolt in the early 1990s, you had a reasonable chance of being left alone.

But what about the situation today? In a previous post I argued that social scientists’ predictions are under-determined by the facts and over-determined by our theories (old post). Something similar may affect normal people as well as social scientists. The world is an immensely complex place and we tend to assume it to be a far more stable place than it is. We do so because the stability that we take for granted is itself a social product, the result of intersubjective norms, institutions, rules etc. that we reproduce in our daily routines.

Iraq today may be far less bloody and far more wealthy – on average – than it was under Saddam, but it is also far more random. Given such a constant (as opposed to lumpy) level of random violence, old certainties no longer apply, old institutions no longer operate, and old norms are routinely violated. People (Iraqi or American) do not deal well with such environments and try to reduce this uncertainty by acting to protect themselves against it. In doing so they promulgate new norms (usually based on old scripts) such as re-imagining group membership on, for example, ethnic or religious lines, as seems to be happening in Iraq. Doing so may of course increase other agents’ uncertainty and thus ratchet up the violence, for every new ‘in-group’ there has to be a corresponding ‘out-group’, but it is a coping mechanism nonetheless.

Overall then, conditions in Iraq may be ostensibly better today than they were in the past, on average, but they may feel worse, and that’s what counts. Even though the body count is lower, even though there is more electricity, and even if there is more wealth in the country, such factors, and focusing on such factors, may be less important to understanding where Iraq is heading than we think. The Iraqi people “may be better off now then they were under Saddam”, but if it doesn’t feel any better to the people on the ground, we should not expect less bodies and more wealth –- on average –- to really make a difference.