Martin Wolf: The failure to calculate the costs of war
An analysis by Linda Bilmes of Harvard University and the Nobel-laureate Joseph Stiglitz of Columbia University suggests that the administration underestimated the economic costs by considerably more than an order of magnitude.* To paraphrase erstwhile senator Everett Dirksen: “a $100bn error here, a $100bn error there and pretty soon you are talking real money”.
So far the government has spent $251bn in hard cash. But the costs continue. If the US begins to withdraw troops this year, but maintains a diminishing presence for the next five years, the additional cost will be at least $200bn, under what Profs Bilmes and Stiglitz call their “conservative” option. Under their “moderate” one, the cost reaches $271bn, because troops remain until 2015.
Additional costs must be added: medical treatment; cost of injuries; disability payments; cost of demobilisation; need for increased defence spending (partly because of higher recruitment costs in the aftermath of the war) and additional interest on debt. Such costs will be bigger, the longer and greater the troop presence. Under the conservative scenario, the total budgetary cost is estimated at $750bn. Under the moderate scenario, the cost is $1,184bn. To put this in context, the minimum budgetary cost is 10 times the world’s net annual official development assistance to all developing countries.
Now consider the wider costs to the US economy. In the conservative case, the adjustments add $187bn to the budgetary cost, even if the macroeconomic impact is ignored. In the moderate one, they add $305bn.
What are these economic costs? The difference between the wages reserves earn in their normal occupations and the lower wages they earn in service are a cost. While life is priceless, the government necessarily values lives in making its decisions. Using the Environmental Protection Agency’s valuation of $6.1m, the authors conclude that the cost of fatalities will be at least $23bn. There are also ongoing economic costs from the terrible injuries. Finally, there is accelerated depreciation of military hardware.
So far, then, the economic cost comes out at a minimum of $839bn (excluding interest). This, alas, does not end the story. In one area, at least, further costs are evident: the jump in the price of oil. Mr Lindsey is reported to have said that “the best way to keep oil prices in check is a short, successful war on Iraq”. He was wrong. Oil production in Iraq has plummeted, from around 2.6m barrels a day before the war to 1.1m.
Before the war, the oil price was expected to remain at between $20 and $30 a barrel. In practice, it has been more than twice as high. The authors’ conservative assumption is that $5 of this is due to the war. Their moderate assumption is that the impact has been $10 a barrel. A $5 increase imposes a cost of $25bn a year on the US and a $10 increase one of $50bn.
Higher oil prices have wider macroeconomic effects. In the short to medium run, spending by the now poorer consumers tends to fall faster than spending by the now richer producers rises. Central banks concerned about inflation also adopt tighter monetary policies than they would otherwise do, while fiscal policy does not normally adjust swiftly to such changes. With a modest “income multiplier” of 1.5, the conservative estimate of the additional losses in output is $187bn over five years. With a multiplier of two and the higher price effect, these costs rise to $450bn.
The authors also add two differences between expenditures on the war in Iraq and likely alternatives: first, they are overwhelmingly abroad; second, they do not contribute directly to consumption, either now or in the future. With these costs taken into account, the total macroeconomic costs may add up to $750bn (see chart) and total costs to $1,850bn.
Critics will stress that both authors served under President Bill Clinton. In the current heated atmosphere of US politics that will be enough to discredit their analysis. It should not do so. Whether or not one believes the war was justified, one should still be concerned that a decision to go to war was taken in the absence of any intelligent analysis of the likely costs.
Nor can one argue that it was impossible to do such an analysis. As I pointed out in a questioning column on the war published almost three years ago (this page, February 4 2003), William Nordhaus of Yale had already prepared a superb analysis, which suggested that $100bn was the lowest imaginable cost and close to $2,000bn perfectly conceivable.
The present analysis also ignores a host of significant economic and non-economic effects. Among these are: costs borne by other countries, including those created by higher oil prices; costs consequent upon creating a link between Iraq and the jihadi movement that did not, on the evidence, previously exist; costs of increasing the income of some of the world’s least desirable regimes, above all, Iran’s; costs of throwing away the option to fight ground wars elsewhere or to fight in Iraq later on, under better conditions, better information and a better state of preparedness; costs of enraging many Muslims; costs to the effectiveness of the US military; costs of fragmenting the western alliance; the loss of Iraqi lives; the cost to US credibility of going to war on a false premise; and the cost to the US reputation of the torture scandals.
It is possible to argue that the benefits for Iraq, the Middle East and the world will outweigh all these costs. But that depends on the emergence, in Iraq, of a stable and peaceful democratic order. That has not yet been achieved.
Even those who supported the war must draw two lessons. First, the exercise of military power is far more expensive than many fondly hoped. Second, such policy decisions require a halfway decent analysis of the costs and possible consequences. The administration’s failure to do so was a blunder that will harm the US and the world for years to come.
理清美国伊战的糊涂账
伊拉克战争开始前，时任乔治?W?布什总统经济顾问的劳伦斯?林赛(Lawrence Lindsey)提出，战争开支可能达到2000亿美元。白宫立刻将他解雇。林赛确实错了，但他错在太低估战争开支了。政府估计的约500亿至600亿美元开支不过是幻想，正如萨达姆?侯赛因的大规模杀伤性武器和许多其它事情。
哈佛大学的琳达?比尔米斯(Linda Bilmes)和哥伦比亚大学的诺贝尔奖得主约瑟夫?斯蒂格利茨(Joseph Stiglitz)在一份分析中指出，政府低估了经济成本，程度远不止一个数量级。*用前参议员埃弗里特?德克森(Everett Dirksen)的话说：“这儿一个1000亿美元的差错，那儿一个1000亿美元的差错，很快就是一大笔钱了。”
迄今为止，政府已花费了2510亿美元现金，但开支还在继续上升。按照比尔米斯和斯蒂格利茨教授所说的“保守”选择，即使美国今年开始撤军，但今后5年仍维持逐渐减少的驻军，那么至少还需要2000亿美元开支。按照他们的“温和”估计，开支将达2710亿美元，因为驻军要保持到2015年(见表)。
必须增加的额外支出有：医疗、伤员开支、残废补偿、复员开支、国防开支增加的需要(部分原因是战后征兵成本上升)，以及债务的额外利息。驻军时间越长、规模越大，这种开支就越高。在保守的情况下，总预算开支估计为7500亿美元。在温和的情况下，开支为1.184万亿美元。从大背景来看，最低的预算开支，是全球每年向所有发展中国家提供的净官方发展援助的10倍。
现在，让我们从更广的范围来看看给美国经济造成的代价。保守情况下，即便忽略宏观经济的影响，调整也将令预算开支增加1870亿美元。温和情况下则增加3050亿美元。
这些经济成本是什么？预备役部队从事普通职业所赚的工资，与服役所赚低工资之间存在差价，这就是成本。生命无价，但政府在决策时必然要给生命定价。根据美国环保总署(Environmental Protection Agency)610万美元的估价，两位作者推断，死亡成本将至少为230亿美元，还有重伤人员的经济成本在上升，最后是武器装备的加速折旧。
迄今为止，经济成本最低达8390亿美元(不计利息)。唉，这还没完呢。至少在一个方面，成本在显著上涨，即石油价格的猛涨。据报道，林赛先生曾表示，“控制油价上涨的最佳方法是速战速决对伊战争”。他错了。伊拉克石油产量骤降，从战前每天约260万桶降到了110万桶。
战前，人们预计油价会维持在每桶20美元至30美元的水平。而实际上，油价涨了一倍多。两位作者的保守意见认为，其中5美元的上涨原因在于战争。他们的温和意见则认为，战争的影响为每桶10美元。油价上涨5美元，就会给美国带来一年250亿美元的开支，上涨10美元则需500亿美元开支。
油价上涨对宏观经济的影响范围更广。中短期来看，比较贫困的消费国的支出下降速度，会比比较富裕的产油国的支出上升速度快。担心通胀的各央行也采用更加紧缩的货币政策，否则它们不会这么做的，而财政政策通常不能对这种变化做出迅速调整。以适中的“收益乘数”1.5计算，保守估计，未来5年内产出的额外损失将达1870亿美元。如果乘数为2，而且价格影响更大，那么这些成本会升至4500亿美元。
两位作者还加上了进行伊拉克战争与其它可能手段之间的两个支出差额：首先，它们离本土太遥远；其次，无论是现在还是将来，它们都不会对消费做出直接的贡献。考虑到以上这些成本，总的宏观经济成本可能达7500亿美元(见图)，而总成本则达1.85万亿美元。
批评人士会强调，两位作者都曾为比尔?克林顿(Bill Clinton)总统效力，在目前美国炽热的政治气氛下，这足以令他们的分析变得不足取信。但不应该这样看。不管人们是否相信这场战争理由正当，仍然应该关心的问题是，在作出开战决定时，没有对可能的开支进行任何理智的分析。
也不可以说，进行这样的分析是不可能的。在近3年前发表的战争提问专栏中(2003年2月4日，本版)，正如我所指出，耶鲁大学的威廉?诺德豪斯(William Nordhaus)已准备了极好的分析。他在分析中提出，可以想象的最低开支为1000亿美元，而接近2万亿美元也完全可能。
目前的分析也忽略了许多重要的经济和非经济影响，这些影响包括：其它国家承受的代价(包括油价上升造成的代价)；因与伊拉克和圣战运动建立了联系而产生的代价（有证据显示，过去并不存在这种联系）；增加世界上某些最不讨人喜欢的政府（尤其是伊朗政府）的收入而产生的代价；放弃在其它地方进行地面战争，或在有更好的条件和情报、准备工作更充分之后，再进攻伊拉克而产生的代价；激怒许多穆斯林的代价；美国军事效果的代价；分裂西方盟友的代价；伊拉克人的生命损失；美国在错误前提下进行战争，其可信度受到影响的代价；美国的声望因虐俘丑闻而付出的代价。
人们有可能争辩说，伊拉克、中东以及全世界得到的利益，将超过所有上述代价。但这取决于伊拉克能否出现稳定、和平、民主的秩序，而这一目标尚未达到。
即使是那些战争支持者，也必须吸取两个教训。第一，动用武力远比许多天真的期望更昂贵。第二，战争决策需要对成本和可能的后果进行像样的分析。美国政府没有进行像样的分析是个大错，这会在今后数年损害美国和全世界。
