————————————————————————————————

Monthly Discussion

The US Presence in Iraq

As the June-30^th date approaches expectations rise that this date will mark the beginning of US disengagement from Iraq. In this Newsletter we examine the possibility that US engagement in Iraq follows a natural-growth pattern. The thinking behind this assumption is that we are dealing here with a well-defined military endeavor that has a beginning, a growth phase, a maturity phase, decline, and an end, and its evolution is subject to the law of competition (survival of the fittest). In this light, we can compare/contrast it with the older and well-documented Viet-Nam military endeavor.

If we take as a measure the number of US soldiers killed, Exhibit 3 shows that this number has steadily grown from the beginning of the Iraq war but showed signs of slowing down toward the end of February 2004. A natural-growth process up to March 2004 indicates an operation practically completed (around 90% of the S-curve’s ceiling).

Exhibit 3. US soldiers killed in action in Iraq. Two S-curve fits are shown: based on data before mid-March 2004 (yellow line), and based on all data up to mid June 2004 (purple line). The data, reported here monthly, come from “Faces of Valor”.

However, after mid-March the rate of casualties increased significantly. But taking more recent data into account, the end-of-February slowdown takes on the hue of a “fluctuation” and a more important S-curve (in purple) indicates that the process may only be 26% completed by mid-June 2004.

If we now look at the Viet-Nam data, we see that its S-curve requires more than 5 years to become completed, see Exhibit 4. Notice that we have much more statistics here and therefore no fluctuations are visible. The agreement between the data-point pattern and the S-curve is excellent.

Exhibit 4. US soldiers killed in Viet Nam and S-curve fit. The data are reported yearly.

To make a meaningful comparison between Iraq and Viet Nam we need to look at similar timeframes. In Exhibit 5 we show the same time scale—if displaced by almost 38 years—and we project an evolution over the next ten years.

In the lower graph of Exhibit 5, the data points cover only a small segment of the S-curve. This fact reflects on the reliability of the forecast. Iraq’s S-curve, based on such a short historical window, may turn out to be considerably lower (or higher!) than the ceiling of 3,500 victims indicated. Still, there are things we learn.

First, the similarity between the two S-curves of Exhibit 5 argues for a much longer involvement in Iraq than advertised in the media. Casualty numbers should culminate around April 2005 and subside only toward the end of 2007. Second, no matter how uncertain this forecast is, it is safe to expect in Iraq a final number of victims at least an order of magnitude smaller (i.e. a factor of 10) than the number of victims in Viet Nam. All this, of course, if “natural” conditions prevail, and by that I mean conditions like the ones during the last couple of years. A new (different) government is not such a condition. To the extent that this government has been intimately involved in the Iraq war, it constitutes “genetic material” for the growing “species.”

Under a significantly different government the evolution of the lower graph in Exhibit 5 may significantly deviate from the S-shaped pattern indicated.

Comparing Iraq to Viet Nam

Exhibit 5. Viet-Nam victims are shown at the top. Iraq victims are shown at the bottom. The S-curve from Exhibit 3 is redrawn here with a timeframe similar to that of Exhibit 4. The similarity of the two curves is striking, but the level of the ceiling in the lower graph is rather uncertain due to the shortness of the historical window.