The Macroeconomics of Love: A Valentine’s Day Analysis

Who says there’s no romance in macroeconomics? Betsey Stevenson and I are currently working on a paper for a forthcoming Brookings Panel, assessing the relationship between levels of economic development and various measures of subjective well-being.

We are working with an absolutely fabulous data set: the Gallup World Poll. The good folks at Gallup are now surveying people in more than 130 countries every year. And they are asking all sorts of interesting questions about subjective well-being.

In honor of Valentine’s Day, let me give you a sneak peek of our findings on love and economic development.

Justin Wolfers Graph Love

It turns out that love is incredibly democratic, and is as common in poor countries as it is in rich countries. And, encouragingly, about 70 percent of us report feeling a lot of love on any given day. This isn’t as obvious as it may sound, as love’s enemy, anger, is significantly more prevalent in poor countries than in rich.

The data suggest some interesting places to look for love: the Philippines, Rwanda, and Puerto Rico top the “love tables.” Meanwhile those in Armenia, Uzbekistan, and Tajikistan are feeling the least love. (Unfortunately, the love question wasn’t asked in the United States.)

While subjective data are relatively new to economics, I’m very excited by what we can learn, and will be sure to get back to you with more results as our research continues.

Now that we can measure something about Gross National Love, you can be sure that economists will start theorizing about it, and crunching these data to test their theories. What is your theory of love, and how might we test it in the data?


Love matters most whatever the status of the person.


The y-axis of the graph is mislabeled. It is not a percent as is marked, but a fraction, where a value of unity corresponds to 100%.
The plot implies that only 0.7% of people reported feeling a lot of love, but that is really supposed to be 100 times greater, or 70%. Thank goodness it is the latter, not the former! What a depressing impression of the people of the world that would give!
A small but important typo!

Jere A. Stecklein

subjective (well being)? Isn't subjective another way of saying opinion?

Cyril Morong

Sorry, this is the first paragraph of the abstract. I did get the whole thing in

"Romantic love is characterized by a preoccupation with a deliberately restricted set of perceived characteristics in the love object which are viewed as means to some ideal ends. In the process of selecting the set of perceived characteristics and the process of determining the ideal ends, there is also a systematic failure to assess the accuracy of the perceived characteristics and the feasibility of achieving the ideal ends given the selected set of means and other pre-existing ends."

Lyn LeJeune

Subjective well-being = love? Love is experienced in poor countries as well as rich countries, or well, economically developed countries, but the questions were not asked in the United States.
Wow! I wish (sarcastically, I wish!) I were head of a big corporation or the marketing guru because boy could I use this information to sell some pretty crappy stuff to convince the people who live in the countries that were "polled" that they need, let's see, candy, highly sugary food, musci full of violence, canned baby formula....can we go on with this list? And were Americans and other developed countries not asked because most of the citizens have become satisfied along Maslow's heirarchy of needs. Now I'm hoping that this research and information is used by responsible agencies to bring good food, water, medicine, books, and other things to countries and that well-being is well-defined and we can all get on with the business of loving because it is simply part of the human condition as is hate. Sure, when I am not hungry and sick, I can love much better, I can show that love with all that I am, but I still love. Perhaps well-being, well-defined, will help us all love better. Nix the junk sales in the global economy. Love does not = taking advantage and harming those who need the most.

Lyn LeJeune-Rebuilding the Public Libraries of New Orleans, "All Things May Not Be Well, And All Things May Not be Well, And All Manner of Things May Not be Well," at


Joe D

I notice that the former SSRs concentrate at the bottom of the "love" scale, *and* that their trend slope is significantly higher than the overall slope. If these nations are removed from the analysis, how flat does the best fit get? Wouldn't that mean "money can't buy me love" (except in the former Soviet Union)?


Sounds like an interesting study. I hope a great deal of care was taken with the questions in the poll. Considering that the meaning of love can vary widely between cultures, it could be quite difficult to create questions that ensure that all responders are talking about the same thing. This would only be compounded by difficulties arising from the different languages involved. (I'm assuming that with 130 countries participating, there are many languages in use.)


Is it possible that love and similar concepts substitute for material wealth for people who are poor or feel they deserve more than they have, as judged by previous wealth or average wealth in their respective communities? Is that why love seems to be less important in marriage for the truly rich, as least as they are shown to those of us who don't have direct interaction with them? Is there a similarity to religion?

ML Harris

Wow. We found the patently obvious. Hate is inversely correlated with wealth. Which again begs the question of why we give hate a chance with military solutions rather than cooperation and development (and possibly wider use of marijuana).

Here's the real finding: build infrastructure, reduce hate. Roads in the middle east. And power plants. Light rail. Etc. These are the proper tools to wage a war on terror.

Guillaume ARNOULD

"All you need is love"


I tend to agree to with Victor (#5). The assumption being made is that emotions measured here in the US will be synonymous with self reported emotions in other language communities. A quick tour around any anthropology department and a few questions regarding emotions in other areas of the world will relate an interesting fact; emotions are given names, and some emotional names do not have counterparts in other language communities... because the self described concept is different.

Just as an example within Western history... in English, there is one word for "love." In Greek, there are a variety of types of love... from agape to eros.

I'd have to see the survey question that Gallup is using and probably talk to each of the translators on whether the question would even make sense in the individual language community that is being polled. Also, there are 193 recognized countries in the world... and about 245 entities that are either recognized or unrecognized as a country in the world. So 130? Hmm....



In certain cultures, telling other people you feel love and things like that isn't really the norm. In American culture, for example, it would be acceptable to say something like "This is a great party! I love you guys!" but it would be viewed as awkward in Chinese culture. So I think these subjective polls are a little misleading.

Cyril Morong

This is an abstract from an actual journal article

set of perceived characteristics in the love object which are viewed as means to some ideal ends. In the process of selecting the set of perceived characteristics and the process of determining the ideal ends, there is also a systematic failure to assess the accuracy of the perceived characteristics and the feasibility of achieving the ideal ends given the selected set of means and other pre-existing ends.
The study of romantic love can provide insight into the general process of introducing novelty into a system of interacting variables. Novelty, however, is functional only in an open system characterized by uncertainty where the variables have not all been functionally looped and system slacks are readily available to accommodate new things. In a closed system where all the objective functions and variables must be compatible to achieve stability and viability, adjustments in the value of some variables through romantic idealization may be dysfunctional if they represent merely residual responses to the creative combination of the variables in the open sub-system."

The author was K. K. Fung of the Department of Economics, Memphis State University, Memphis. It was from a journal article in 1979. More info on it is at this link. The entire article, which is not too long, can be found at this link.


Tom Selling

Reporting that you feel something and actually feeling it are two different things.

I thought freakonomics was about studying what peole *actually* do (like real estate agents when they sell their own house), and not relying to any degree on what people *say* they do.

Marc Robinson

I have two theories:
1) Love is positively correlated with freedom, particularly non-political freedom (so China under Mao, Soviet Union under Stalin, and Iran under Khomeini would score low), with the effects of totalitarian state long-lasting. (Test with measure of freedom (Freedom House?) both current and 25 years ago.)
2) Love is correlated with Western traditions of romantic marriages rather than arranged or convenience marriages (requires cultural knowledge of frequency of arranged marriages; note, it will be interesting even if hypothesis is proven wrong).


I agree with a number of the comments that cultural differences make getting comparable data difficult. Unlike them, though, I expect that the Gallup Pollsters thought of that as well and crafted their questions with those differences in mind. If not, then they should have their polling merit badges revoked.

Nick Fortescue

I'm sceptical like a lot of the other comments but for a different reason. How do you control for language? Love is a notoriously difficult word to translate, and laden with cultural differences. Ask any Bible translator.


Can you really measure love like you measure tangible economic objects like GDP or money? If you have a dollar, you have a dollar. If you are in a good mood and you have a dollar, you still have a dollar. If you are in a bad mood and you have a dollar, you still have a dollar. A dollar no matter what is 100 pennies. When it comes to love it is not as clear cut. If a person is having a bad day it is not unlikely that they do not feel that they have a lot of "love." It is also very hard to define a word such as love, especially in different countries. You can say a dollar translates into so many euro's but how can you translate the meaning of love from one country to another? Is love thought to be the same thing in countries where there are things like arranged marriages? I don't believe that this research will prove useful because i don't believe that it will prove accurate.


What a bunch of nonsense! Please don't pretend that this is science in any form. Why would you even expect to find anything unless you have done absolutely no background reading on cultures, or have absolutely no experience outside your own narrow demographic? That this would pass as academic research is mind boggling.


I agree with Victor and the seconder who point out that this has more to do with culture than actual differences in our state of happiness. Not only are some cultures reticent, some value urbane cynicism and regard a big smile as evidence of stupidity. You ask some citizens of the former Soviet bloc if they "felt love" and you might as well be asking them if they feel like admitting to moments of mindless crotch-fondling.