Friday, March 17, 2006

Cost of the Iraq war

In 2002, Lawrence Lindsey, director of the White House National Economic Council, surprised the Bush administration by saying that a war in Iraq could cost $100-$200 billion. Some considered his estimate to be exaggerated, but it is now seen as far too low. Even accounting for costs already incurred alone these figures are already exceeded (estimates of costs are being updated by the second here) .

Various economists are now working to get a feel for the scale of the costs.

(i) Scott Wallsten ‘The Economic Cost of the Iraq War’ here. At this website you can make your own costing assumptions to come up with your own cost estimate. Estimated sunk costs $500 billion with $500 billion yet to come – ignores possible benefits. The simulation exercise is fascinating, simple and grim.
(ii) Joseph Stiglitz ‘The High Cost of the War in Iraq’ here. A popular version of (iii). Emphasises the cost of the war in terms of foregone social and other programs.
(iii) Joseph Stiglitz and Linda Bilmes, ‘The Economic Costs of the Iraq War’ here. Costs exceed $1trillion – again ignores possible benefits
(iv) S. J. Davis, K.M. Murphy & R.H. Topel, ‘War in Iraq Versus Containment’ here. This 2006 study comes up with cost estimates in 2003 dollars of $410-$630 billion. This is then compared to the cost of maintaining the prior ‘containment’ policy which they estimate at $300 billion. But the ‘containment policy’, they claim, was already being criticized as ineffective and, allowing for possible costs here, total costs of containment turn out to equal the cost of intervention. The authors see net economic benefits to the people from the war in Iraq which tilts the scales in favor of intervention. An earlier study by these authors (here) dated March 2003 seems dated by subsequent events.

I'll try to collate further studies as I become aware of them. I think (iv) is the most methodologically sound approach but, its empirical assumptions have been questioned. Tyler Cowen at Marginal Revolution doubts that a Saddam-led Iraq would have raised 'the probability of a major terrorist attack by 4 percentage points in any given year...'. On the other hand, Cowen argues, perhaps the current civil war might have occurred, sooner or later, if we had stayed out. There is also a good discussion at the Becker-Posner blog here.

I set out the Davis et al conclusions below.

Prior to the invasion of Iraq a containment policy was the leading alternative to forcible regime change. Davis et al. ask:

· Was war more or less costly for the US than continued containment?
· Would a continuation of containment have saved Iraqi lives?
· Is war likely to improve the economic well-being of Iraqis?

There answers are ‘no’, ‘no’ and ‘yes’ respectively to these questions. I summarise their conclusions.

Forcible regime change in Iraq has been costly. At January 2006, the Iraq intervention implies present value costs for the US of $410-630 billion at a 2% discount rate. This captures the estimated economic costs of US military resources, the value of lost lives and injuries by US soldiers, lifetime medical costs of treating injured soldiers, and outlays for humanitarian assistance and postwar reconstruction. Pre-invasion views about the likely course of the Iraq intervention imply present value costs in the range of $100-$870 billion. Military resources devoted to postwar occupation account for more than half of total costs except in optimistic scenarios that envision a short occupation, little postwar conflict and a smooth reconstruction effort.

The view that the high cost of the intervention is an argument against the decision to invade is deficient because it ignores the costs of alternatives. Containment costs were likely to be $300 billion even if it would be completely effective in achieving its national security goals.

Moreover advocates of forcible regime change expressed concerns about containment. To evaluate these concerns, the possibility that effective containment might require the mounting of costly threats and might lead to limited or a full-scale regime-changing war at a later date is modeled. Also the possibility that the survival of a hostile Iraqi regime raised the probability of a major terrorist attack on the US is considered. These contingencies sharply raise the cost of containment. Factoring these contingencies in yields present value costs for containment from $350-$700 billion. These figures are in the same ballpark as the likely costs of the Iraq intervention from the vantage point of early 2006. Thus, even with the benefit of partial hindsight, it is difficult to gauge whether the Iraq intervention is more costly than containment.

Davis et al also consider the consequences of the war-versus containment choice on the economic well-being of Iraqis and the loss of Iraqi lives. They conclude the war will lead to large improvements in the economic well-being of most Iraqis relative to their prospects under containment. This follows because the Iraqi economy was in terrible state before the war, and it would have remained so under containment. The Saddam regime caused real income per capita to fall 75% as a consequence of misrule. Much of Iraq’s greatly diminished output was diverted to an oversized military, an apparatus of terror and repression and the relentless glorification of Saddam. Finally, the removal of sanctions, the expansion of petroleum exports, large-scale reconstruction aid, and the reintegration of Iraq’s economy into the world economy provide a strong basis for economic gains – even in a society with serious institutional weaknesses. If, over the course of a generation, Iraqis recover half the economic losses suffered under Saddam, they will be significantly better off in material terms as a consequence of regime change.

The regime also brought torture, repression, displacement and death to huge numbers of Iraqis and others. The regime killed more than 500,000 Iraqis. Under containment after the 1991 Gulf War, at least 200,000 Iraqis died at the hands of the regime or as a direct consequence of its policies, including its refusal to comply with UN Resolutions and its diversion of oil revenues and other resources to palaces and monuments. Had containment remained, history suggests that premature Iraqi deaths would have continued at 10,000-30,000 per year. The weight of evidence points to a greater Iraqi death toll from containment. Perhaps the strongest reason to question this assessment is the possibility that a post-war Iraq could devolve into an extended and large-scale civil war. This possibility cannot be ruled out. What can be ruled out in light of the evidence is that the leading alternative to war involved little loss of Iraqi lives.

No comments: