What’s in a name?

Economists analyse the oft-quoted remark on the ironic significance of names in the famous soliloquy  “What’s in a name … a rose by any other name would smell as sweet”. This is a poetic illustration of the argument that it is not the name of things that matter but rather what they actually are. Whilst it is true that a rose would emit the same volatilised chemical compounds if it were named something else this does not necessarily mean names do not matter.

Take the labour market for instance. Would equally qualified employees with different names receive the same or different treatment when applying for jobs? Furthermore, does this mean there are economic costs and benefits associated with certain names?

It is hardly controversial to say that names, just like prices, convey signals. They can provide information on gender, race, class and more. So it is not unreasonable to assume discrimination in the labour market may occur as a result of names.

A famous study by economists Marianne Bertrand and Sendhil Mullainathan entitled “Are Emily and Greg More Employable than Lakisha and Jamal?” argues that African-American sounding names receive worse treatment in the labour market than white-sounding names.

Bertrand and Mullainathan conducted a controlled experiment where they sent out nearly 5000 identical-quality resumes attached with either African-American or white sounding names in response to 1300 advertisements across Boston and Chicago. The results showed that applicants with African-American sounding names had to send out around 15 resumes to get a callback whereas those with white-sounding names only had to send out 10. This variation in callback rates was statistically significant.

Another famous study by economists Roland Fryer and Steven Levitt entitled “The Causes and Consequences of Distinctively Black Names” offers a seemingly contrasting view. Fryer and Levitt found that after controlling for negative economic conditions at birth the effect of a name alone on life outcomes was substantially and statistically insignificant.

They argued that a distinctively African-American name typically correlates with lower socioeconomic status (SES) at birth and it is the lower SES background and not the name that has an effect on the person’s life. It has been suggested that this argument explains the results of Bertrand and Mullainathan’s study. The logic is as follows. An African-American sounding name signals lower SES, on average. Employers believe lower SES hinders human capital accumulation, which equates to lower productivity. Accordingly, they might discriminate on that basis rather than due to animosity towards a particular race.

In an analysis of these studies by Kristie M. Engemann and Michael T. Owyang, it was concluded, “ethnic names appear to serve as a hindrance in the labour market, but the exact extent has yet to be conclusively determined.”

The discrimination against certain names in the labour market is not necessarily limited to the racial information they convey. Another 5 studies concluded that easy-to-pronounce names are generally evaluated more positively than difficult-to-pronounce names – bad news for me.

Even if there is some discrimination in the labour market against names it does not necessarily mean a name has an economic cost associated with it unless the discrimination leads to adverse outcomes in the long run. This is very much what Fryer and Levitt argue. They argue that any discrimination that exists is not substantial or powerful enough to cause any long-run detriment.

Whether it is on the basis of race, SES, how pronounceable it is or some other factor there appears to be discrimination against names in the labour market. Whether this has an impact or not on eventual economic outcomes is where the question becomes debatable.

It is at this point I should declare that I have a vested interest in this question. I currently go by my Greek name Yannis. However, the first name listed on my birth certificate and other legal documents is John (the anglicised version of Yannis). If an ethnic or difficult-to-pronounce name has an economic cost associated with it then, on a cost/benefits analysis, I should ditch the name Yannis in favour of John.

Perhaps a simple CBA is not the best way to evaluate the choice of a name though. Consumer choice theory would state that costs and benefits are not the only variables in the consumer’s utility equation. Preferences are also part of the equation. For example, people sometimes have a preference for a particular brand of confectionary and this preference might be so strong that they choose to buy it over a cheaper substitute. That consumer may have achieved equilibrium between preferences and consumption costs. This can apply for   names too. If the economic cost associated with a name is low (as Fryer and Levitt would argue it is in the long-run) someone might choose a potentially costlier name due to a preference for it.

In my case, I think I like the name Yannis better than John (no offence to anyone named John). However, quantifying preferences and utility is notoriously difficult if not impossible (ordinal utility theory states that it is impossible). The question remains, does the possible economic cost associated with an ethnic or difficult-to-pronounce name trump my preference for that name or vice versa? My analysis thus far has failed to answer this pertinent question so for now I shall adopt Juliet’s mindset and not fret about names.