Demographic Change: How the U.S. is Coping with Aging, Immigration, and Other Challenges

Date April 9, 2007
Speaker William H. FREY(Research Professor, Population Studies Center, University of Michigan / Senior Fellow, Demographic Studies, Milken Institute / Visiting Fellow, The Brookings Institute)
Moderator NAKATA Daigo(Fellow, RIETI)


There are three major changes happening in the United States that have particular relevance to the demographic situation in Japan. The first is the aging of the population, due in large part to the baby boom population getting older. The second is the increasing number of immigrants coming to the U.S., contributing to population growth and slightly counter-balancing the aging of the population. The third is the increase in racial and ethnic diversity, in particular an increased proportion of Hispanic and Asian populations, which raises social integration issues. These three changes interact with each other in ways that affect policy discussions about the older population, social integration, and the labor force.

To provide some background to these phenomena, the U.S. has a fertility rate of 2.05, which means that it is almost able to renew itself, and is significantly higher than many other developed countries, particularly Japan, while life expectancy is somewhat lower. Immigration is at a comparatively high rate, while Japan's is very low.

The population of the U.S. is projected to grow by 40% between 2006 and 2050, while other countries will either grow at a much slower rate or decline due to their lower fertility and immigration rates.

Around 12% of the U.S. population is aged over 65. Within about five years the growth of the U.S. labor force will start to decline, due to the baby boomers retiring, although the growth is still positive. In contrast, around 20% of the Japanese population is aged over 65, and there is negative growth in the labor force due to the low immigration and fertility rates.

Age dependency is a measure of the number of people aged over 65 as a percentage of the labor force. Between 2010 and 2050 the U.S. will go from an age dependency below 20% to over 30%, while Japan will go to over 70%. Even though the U.S. is not aging as fast as Japan, it is aging nevertheless, causing a great deal of concern.

In the current decade between 2000 and 2010 the biggest population growth is in the baby boomers - people aged 55-64 - followed by people aged 45-54. On the other hand, the population aged 35-44 shows a decline, because the smaller generation going into that age group is not replacing those baby boomers moving up into the older group. Projecting that ahead for the next three decades, the baby boomers will move into the 65-74 age group between 2010 and 2020, and into the 75-plus group between 2020 and 2030. During that period we need to be concerned about rising health care and social security costs. At the same time there will be fewer people in the high-income age group to provide for those costs. This is the situation in the U.S. for the next 20 or 30 years, and it is causing a lot of concern.

Baby boomers are very different from their parents. They have fewer children, are less likely to have stayed married, are better educated, and have many more women in the labor force. Therefore when they become seniors they will not have as steady and stable a family structure. In addition, studies have shown that the baby boom generation is much more socially unequal within the generation in terms of income. So while some will live very luxurious retirements, a significant portion of that baby boom population will struggle financially, will not have private pensions, will need to continue working. A large proportion of these will be women living alone, and the most likely to be poor and vulnerable.

Although this is the situation across the U.S. as a whole, there is substantial variation between regions within the U.S. The young population is moving away from the northern and interior parts of the country. This means the northern and central states are going to have a much more severe age structure, and even today the percentage of the population aged 65 and over is much higher in those central states than in other parts of the country. That means there is not only a national aging issue affecting federal programs such as social security and Medicaid, but also state expenditures on support services for senior citizens are going to be much more difficult in those states where there are not as many young people to prop up the senior population as it starts to get older.

Projections show that by 2020 health care and social security allocation in the U.S. federal budget will be 91% of the projected federal revenue. By 2030 it will be 121% of revenue, which is clearly impossible. Spending on seniors will have to be adjusted, and some hard choices will need to be made. People may have to work longer or take fewer benefits, or everyone else will have to pay higher taxes. Keeping benefits and retirement age at the current level would result in prohibitively high taxes for the relatively small younger generations.

This is why people are starting to look at immigration as a possible way to increase the size of the younger population. After 1970, as a result of a fundamental change in the immigration law in 1965, the immigration rate shot up, and is now at an all-time high, with 35 million immigrants making up 12% of the population.

However, there are also some problems associated with immigration. 30% of the foreign-born population is referred to as "undocumented," having come to the U.S. illegally, either by crossing the border without permission or coming with temporary permission and simply staying in the country. A very large part of this undocumented population is made up of unskilled workers, many of them from Latin America. About 78% of the undocumented immigrants and 41% of the legal immigrants come from Latin America, while a large proportion of the legal immigrants come from Asia. This is generating some tension and culture clash in the U.S.

Of the legal immigrants, around two-thirds do not come to the U.S. for work purposes but rather because they have family relationships with other immigrants. This is an implication of the 1965 immigration law, which says that people wishing to immigrate to the U.S. have a high priority if they have family already living there. About 19% of immigrants come to the U.S. for employment reasons, and 17% for other reasons such as refuge or asylum.

One positive aspect of immigration is that many immigrants are coming to the U.S. during their working-age years, resulting in a younger population structure. On the other hand a controversial aspect of immigration is that it does not necessarily make the population better educated. Immigrants generally have a lower level of education than native U.S. citizens, although there are also a lot of immigrants with postgraduate education. It is primarily the Mexican and Latin American immigrants who tend to keep the overall immigrant educational level low, many of them coming for family reunification purposes.

The percentage of Hispanics with less than a high school education goes down with the second-generation, but is still more than twice that of the overall U.S. population. Second-generation Hispanics have a higher percentage of college graduates than the first generation, but it is still far below the overall U.S. population. Conversely, Asians have an even higher level of education than the overall U.S. population.

This raises a lot of questions about immigration policy. Should family reunification be restricted? Should undocumented immigrants be sent back to their countries of origin? There is a powerful immigration debate going on. With 12 million undocumented immigrants in the U.S. they cannot simply be sent back to their countries of origin. Rather, one opinion is that they need to be brought into the legal system and mainstream economy, allowing them to be regulated and assisted.

A study of the long-term impact of immigration on the fiscal situation of the U.S. showed that it is related to the level of education of the immigrants. For immigrants who do not have a high-school education, the overall loss to the country is $89,000 per first-generation immigrant, with a gain of $76,000 per person for the second generation, who are able to contribute a little more in taxes, producing a small overall net loss. For those with some college or more education, both the first and second generation contribute to the economy more than they take away. This study shows that immigrants make a small positive contribution if you take into account their second or third generations, although what they do immediately for the labor force is open to question, and more research needs to be done in this area.

Over the past decade the immigrant population has been spreading out to the middle part of the U.S. because there are low-skilled jobs available there. This is causing culture clashes with the existing populations which have had little exposure to non-native U.S. citizens, and are worried that the immigrants will have a negative impact on their community. This is another reason the immigration debate has become much more heated politically, and the U.S. Congress is starting to look more seriously at changes in immigration law. The immigrants arriving in non-traditional locations in the U.S. are not representative of the overall immigrant population: they are less likely to speak English well, more likely to be unskilled, and more likely to be undocumented. This reinforces the negative reaction towards immigration.

In the 1970s, 80% of the U.S. population was white, with the major minority group being African Americans. Hispanics are now the major minority group, and the white population is only 67% of the total. Projecting the current trend forward, by 2050 only half of the population will be white and 20% will be Hispanic. Different areas of the U.S. have different racial and age group compositions, with Hispanics concentrated in the south and west, and the white population concentrated in the north and Midwest. Some areas are therefore much more able than others to be "melting pots" for the new immigrant groups.

For a large part of the country an increasing proportion of the population, in particular children, speak Spanish at home. Typically the children of first-generation Hispanic immigrants speak Spanish at home and English outside the home. However the continuing waves of immigrants mean that Spanish will be a common language in large parts of the U.S. that receive lots of immigrants.

Within these melting pot areas, due to the skill level of the immigrants most professionals or managers are white, while many unskilled or service workers are Hispanic, leading to income level segmentation and compounding the integration problem. Therefore a high priority is to ensure that second and third generations of the Hispanic population receive a good level of education.

The under-15 age group in the U.S. is becoming much more racially diverse, while the older age groups are still very white. There is therefore not only a gap between the economic interests of these two groups but also the racial and cultural interests. This means that further integration needs to be encouraged, and immigration policy might need to be rethought to put a little more emphasis on skills rather than family reunification.

In conclusion, there are a number of policy questions raised by these trends. Should expenditure on programs for seniors be cut back? Should the retirement age be raised? Should taxes be increased? Should immigration policy be more economically strategic, geared towards filling specific job niches within the country? How should illegal immigrants be brought into the system, and provided with worker rights they do not necessarily receive at present? These are difficult issues, but I believe that most people in the U.S. think that immigration helps to make the population more vital, more vibrant, and helps to strengthen the labor force.

Questions and Answers

Q: Is there a large margin of error in the projections?

A: Immigration levels are hard to predict accurately, leading to inaccuracies even in short-term projections. On the other hand, the projections rest on fairly rigorous assumptions. These projections allow us to understand in a somewhat intelligent way what the impact will be of the changing components.

Q: Of the Asian immigrants, which regions within Asia do immigrants originate from?

A: Around 32% of legal immigrants are Asian, many of them coming from China, India, and Korea. We also have temporary labor coming to the U.S., such as IT workers from India. In that sense it is difficult to say how many people are coming from Asia.

This underlines a problem with the immigration law, where temporary measures supplement the law. A more structured immigration law, integrating temporary work visas and based on skill levels, would probably lead to more people coming from Asia and perhaps some European countries, as well as from Latin America.

Q: Will the demographic changes cause shifts in political preferences in regions of the U.S.?

A: There are U.S. states that traditionally vote for the Republican Party and those which vote for the Democratic Party. Some of those states are aging, and some of them are getting younger. Although immigration and social security issues do not cut strictly across party lines, those states getting older are going to be much more in favor of preserving the basis for old-age retirement and old-age health care, whereas the younger states will be more concerned with other issues, and so may cause changes in the traditional voting preferences.

Q: Is the declining fertility rate among the white population a policy issue for the U.S.?

A: It is true that white fertility is below the replacement level, at around 1.4 or 1.5, while Hispanic fertility is at around three, although the longer the immigrant group is in the U.S., the lower the fertility rate.

While some anti-immigrant advocacy groups do express concern about the ethnic composition of the births, I do not think people are actually concerned about whether the new babies are going to be Hispanic or white or any other race; they are simply worried how many babies there will be. There is also a degree of intermarriage among Hispanics and particularly Asians in the younger generation, which is not large but is increasing.

Q: The labor force participation among U.S. women peaked around the year 2000 and has been leveling off since then. Does this imply changes in fertility rate or birth rate in the future?

A: It may simply imply that women at a certain stage in their life would rather move from full-time to part-time work in order to have a more balanced lifestyle. This may lead to an increase in the birthrate, but I do not think this will necessarily be the case.

Q: Japan is a very racially homogenous country. Do you have any suggestion on how Japan might address this?

A: There are probably a lot of government programs and economic incentives that can be done in Japan in order to mitigate the labor shortage and the social pension programs, such as raising taxes and encouraging people to work longer.

In addressing Japan's declining population a simple answer would be immigration, and without a tradition of immigration Japan could start with a clean slate. However immigration also brings with it many problems that have to be dealt with.

Q: Is there any debate in the current presidential campaign on the social security reforms President Bush has proposed?

A: Politicians in the U.S. are afraid to talk about social security, because studies have shown that older people vote at a much higher rate than anybody else in the population. One of the reasons President Bush's proposal was not very well accepted was because the people in Congress were worried about not getting reelected if they voted too heavily toward what President Bush would call a "personal accounts" rather than a "pay-as-you-go" system. So it is a very delicate political subject, and would be a very important subject to have in a presidential campaign because we are going to encounter these problems in the future.

Q: How will the demographic changes in the U.S. affect the U.S. and world economies?

A: Many people concerned with the U.S.' long-term demographic changes point out that the baby boomers are probably the most educated generation, even in comparison to generations subsequent to them, and as they retire this leaves a labor force which has got to fill a lot of niches in a global economy that rewards knowledge-based jobs requiring high levels of education and specialized skills. To the extent that we have a somewhat open immigration policy we can bring in the knowledge and skills needed to replace the baby boomers. I believe the open exchange of people and the open exchange of goods goes hand in hand, so the fact that our immigration policy is open is an important asset.

We do have to think more strategically about the immigration policy. A lot of groups who have looked at our current immigration policy contrast the retirement of educated and skilled baby boomers to the much more socially unequal immigrant group. This will require us to address the influx of unskilled people to the U.S. while putting emphasis on our educational system to ensure that second- and third-generation immigrants are able to move into a knowledge-based economy. If we do that successfully I believe we will be a big player in the global economy, much more than other countries which do not have an open immigration policy.

Q: What is the fundamental reason behind the falling worldwide birth rate?

A: One reason is the education and empowerment of women, and women's independence and participation in the labor force. There is a strong correlation between women going into the labor force and getting better education and lowering fertility. That has occurred in the industrialized world, and has been particularly dramatic in the last 15 years. There has also been easier access to and more diffusion of information in other parts of the world showing that having children is expensive and having children too early cuts opportunities for economic advancement.

*This summary was compiled by RIETI Editorial staff.