Mexican immigrants to the US enjoy better health than their US-born neighbours but immigrants who marry US natives see these healthy lifestyles deteriorate. This suggests that Mexican culture is somehow healthier than that of the US. These are the surprising findings of research by Osea Giuntella, presented at the Royal Economic Society’s 2013 annual conference.

The study examines the so-called ‘Hispanic health paradox’ whereby first-generation Mexican immigrants have better health than US natives even though they are poorer and Mexico has a worse health system. At the same time, second-generation Mexicans have worse health than the first generation even though they are richer.

Some argue that this paradox is due to first-generation immigrants being healthier than average and that this effect subsides after a generation. Others argue that second-generation immigrants begin to lead unhealthy ‘American’ lifestyles.

The study instead argues that the paradox should be reversed. If either explanation were correct, the author claims, then second-generation Mexican immigrants should be even less healthy than they are. But by looking at birth weight as a measure of health, the research finds that there is something in Mexican culture that keeps their health from declining further.

The study shows that first-generation immigrant women are less likely to make unhealthy decisions, such as smoking and drinking during pregnancy. Although such risky behaviour is more likely among second-generation Hispanics, it is still far less likely than for white US natives. The author argues that this persistence of healthier lifestyles explains 76% of the difference in health between second-generation Mexican immigrants and Americans.

The research also finds that Hispanic women who marry US natives lead far less healthy lives than their sisters who marry other Hispanics. Third-generation children of Hispanic-US couples are 14% more likely to be underweight than children from two Hispanic parents. This result goes against the commonly held view that intermarriage leads to healthier and more stable families.

The author argues that this finding suggests the explanations for differences in health across ethnic groups cannot be explained by a ‘pure mechanical statistical process’ and that culture and behaviour play a key role, adding that policies aimed at ‘maintaining healthy behaviours can significantly affect these health patterns’.


Surprisingly, first-generation Mexican immigrants to the US have better health than natives even though they are poorer and Mexico has a worse health system. At the same time, second-generation Mexicans have worse outcomes than the first-generation even though they are richer.

Previous studies refer to these stylised facts as the ‘Hispanic health paradox, which has been observed across several health outcomes (life expectancy, mental health, health status, infant health, etc.). This new study analyses the Hispanic paradox in birth outcomes – children of first-generation Mexican women have better birth outcomes (or example, lower incidence of low birth weight), but this advantage fades away in the following generation – and attempts to shed light on the possible mechanisms behind it.

Possible explanations: selection or behaviours?

Some people argue that this apparent paradox is explained by the fact that healthy Mexicans are more likely to migrate to the US and that the second-generation result is due to the relatively weak intergenerational correlation in health and the fact that the initial selection effect fades away over time (regression towards the mean). Others argue that second-generation Mexicans become more Americanised and have less healthy behaviours than their parents did.

This research shows that the explanation is more complicated. The first-generation difference can be explained by selection, but if the only factor were selection (and the fact that the selection effect fades away across generation), second-generation Mexicans would be less healthy than what is observed. They would be less healthy even accounting for their increased incomes.

A model of selection and intergenerational transmission of health not only explains the apparent paradox in birth outcomes, but it reverses the paradox. In other words, the puzzle is not that immigrants’ relative health deteriorates so rapidly, but that it does not deteriorate rapidly enough.

But accounting for the relative high persistence in healthy behaviours among second-generation Mexican women makes it possible to fit the facts almost perfectly. In particular, this study shows that first-generation immigrant women have substantially lower incidence of both risky behaviours (such as smoking and alcohol consumption during pregnancy) and health risk factors (gestational hypertension) that are known to seriously affect birth outcomes.

Although risk-factor behaviour worsens between first- and second-generation Hispanics, immigrants maintain a sizeable advantage in terms of lower incidence of health risk factors compared with white natives. The persistence of healthier risk-factor behaviour explains 76% of the ‘reverse paradox’.

Cultural assimilation: intermarriage, behaviours and health – a paradox in the paradox

The convergence in birth outcomes is more marked among those who are less likely to maintain the health-protective behaviours and conditions (such as low rates of smoking, alcohol consumption and hypertension) that characterise the first-generation immigrants and, more generally, among those who are more likely to have culturally assimilated.

Interestingly, among second-generation Hispanics, those married within the ethnic group show higher resilience in healthy behaviours. Using ethnic intermarriage (being married to a non-Hispanic white man) as a metric of cultural assimilation, the study shows that third-generation children of intermarried Hispanic couples are 14% more likely to be of low birth weight than children of intra-married couples.

This result is particularly striking because intermarriage is usually associated with positive socio-economic outcomes and is partly explained by the higher incidence of risk factors (smoking, etc.) observed among intermarried women. These results hold when focusing on a sub-sample of second-generation sisters and using within family variation in the patterns of assimilation to explain differences in health convergence.

Is this health deterioration inevitable?

As a whole, these findings show that the health trajectories observed among Hispanic descendants cannot be entirely explained by a pure mechanical statistical process. Socio-economic and behavioural factors mediate the transmission of health across generations. Policies aimed at reducing disparities in access to and quality of health care, and at maintaining healthy behaviours can significantly affect these health patterns.



Osea Giuntella: +1 (617) 817-5665 (

RES media consultant Romesh Vaitilingam:
+44 (0) 7768 661095

Page Options