Making leaders

Dwight D. Eisenhower described leadership as “the art of getting someone else to do something you want done because he wants to do it.” Eisenhower was a successful wartime general and president. What made him successful? It was not a full head of hair and a fit physique, two of the physical traits of a CEO. What made him an unsuccessful university president? Was it luck or skill, or his social interactions with those he led?

There are many theories on what makes a leader effective, where effective leaders go, and whether leaders are born or created, but little empirical work. We cannot run a field experiment to study leadership of an organization in a high-stakes setting. Empirical tests of leadership theories have to come from quantitative studies of leaders and organizations, but large, longitudinal datasets on CEOs and companies are rare. A unique longitudinal data set on Union Army soldiers augmented with information on the regiment level provides a testing ground for leadership theories. This sample (available at uadata.org), created from men’s army records and linked to their census records, is the most comprehensive longitudinal database in economic history. It has been used to study the economics of aging (Costa 1998) and the role that social capital plays in people’s decisions (Costa and Kahn 2008). The data contain information on men’s promotions and demotions, their jobs during the war, their socioeconomic and demographic information at enlistment, and their jobs and locations after the war.

Who became a leader and what made leaders effective? The more able, i.e. the literate and men who were in higher status occupations, were more likely to become officers. So were the tall and the native-born. There were benefits to being an officer – higher pay and lower odds of death, both on and off the battlefield. Game theoretic models of leader effectiveness have emphasized that one way to elicit effort from followers is to lead by example. Although on average commissioned officers did not imperil themselves in battle, when they did, it was an effective strategy in creating a cohesive fighting unit. Company desertion rates were lower for companies in which the regimental battlefield mortality of commissioned officers relative to enlisted men was higher.

After the war leaders moved to where their talent would have the highest pay-off, as predicted by economic models of sorting. The former sergeants and commissioned officers were more likely than privates to move to larger cities which provided higher wages and greater diversity in the dominant economic activity of the time, manufacturing. Even men who started in low status occupations in cities were able to climb the occupational ladder.

Are leaders created or born? The Army, and a large management literature, stresses that leaders have character, presence, and intellectual capacity. In contrast, economic theory emphasizes the management skills that can be learned. Union Army soldiers who missed being promoted because casualty rates were relatively low in their companies were less likely to be in a large city after the war compared to men who were promoted, suggesting that in the long-run leaders are created. One of the skills learned in the army may have been to be a generalist. Sergeants and commissioned officers with more than strict military tasks while in the army were more likely to be in large cities.

A Civil War context for testing theories of personnel economics may be unusual. The 150th anniversary of the Civil War has focused more on historical research and re-enactments. But if a stress test of theories is their explanatory ability in very different contexts, academic personnel economics does very well.

Headline image credit: Field Band of 2nd R.I. Infantry. Photo by Mathew Brady. War Department. Office of the Chief Signal Officer. Brady National Photographic Art Gallery. US National Archives and Records Administration. Public domain via Wikimedia Commons.