Decline of Paired-Adult Households, and the Rise of Single-Adult Households

My research in household finances suggests that marriage is a major determinant of households’ financial fortunes.  Married households generally do much better financially, because they have the potential for multiple income streams and the ability to share costs.  This week’s posts probe the relationship between changing living arrangements and households finances.  In this post, we look at the rising incidence of living outside of paired-unions.

Decline Paired Unions Since 1967

The proportion of households headed by adults in a paired union has declined markedly since 1970. This change is depicted in the figure below, which is based on Census data.1

The proportion of married adults has declined substantially, while the incidence of cohabitation has grown dramatically. Rising cohabitation rates have not fully offset the decline in marriage. In 1967, 70.3% of US adults lived with a spouse, 0.4% lived with a non-spousal partner, and about 7.6% lived alone. By 2015, 51.4% lived with a spouse, 7.5% lived with a non-spousal partner, and 14.4% lived alone. This leaves a substantially smaller proportion of adults living outside a paired union (70.7% in 1967 vs. 58.9% in 2015, a difference of 11.8 percentage points).

Much of this change has produced a rise in adults living alone, which doubled from 7.6% of households in 1967 to 14.4% in 2015.

The remaining adult population appears to have shifted to alternative cohabitation arrangements with non-partners. Other living arrangements, like living with other relatives (9.8% in 1967 to 12.1% in 2015) or non-relatives (1.3% to 3.6%) rose, but the incidence of living with one’s parents is roughly the same as it was decades ago (11% in 2015 vs. 10.6% in 1967)

Do these changes have financial consequences?  There are many indications that the answer is yes.  Marital status appears to have a significant impact on households’ incomes and net worth.  In my forthcoming book, I argue that being in a paired union sways a family’s ability to stay in the middle class.  The consequences of being single while having children seem particularly strong, because children require considerable investments of labor that either have to be met by foregoing work or using commercial alternatives.

This does not necessarily imply that staying in a first marriage is critical.  Many of these benefits may exist in unmarried cohabitants or recombined families (in which previously divorced adults recombine into a new paired union household).  They may also exist in other multiadult household arrangements, like living with a sibling or roommate.  These questions will be examined in future posts.


  1. Census Bureau (2015) “Table AD-3. Living arrangements of adults 18 and over, 1967 to present” Data table from Census Bureau.http://www.census.gov/hhes/families/data/adults.html

The Rise and Fall of Unionism in America

America’s workforce was most heavily unionized in the early 1950s. Since then, unions have experienced a long-term decline.

Unions are a persistent point of conflict in economic policy debates. Many observers argue that the economic difficulties faced by America’s middle class are at least partly attributable to the decline of unions. Others see unions as a detrimental force in the economy, and believe that they harm employers, the US economy, and ultimately workers themselves. Unions are portrayed as powerful and corrupting forces in US society, and dying institutions that are being squashed by all-powerful business interests.

How strong are unions? One way to answer that question is to look at union density, the ratio of unionized workers to total workers. The figure below depicts changes in union density across all US workers from 1880 and 2013. Data come from Klaus Walde and Barry Hirsch and David Macpherson.1

Union Density

The graph suggests that unionization developed slowly between the 1880s and the Great Depression. By the 1880s, struggles to unionize labor were heated and sometimes violent. Larger movements to advance unionism included the Knights of Labor, and later the American Federation of Labor. Through the Great Depression, more workers unionized under the auspices of the Congress of Industrial Organizations. These latter two groups eventually merged to create the AFL-CIO.

Unionization increased dramatically with the institution of the 1935 National Labor Relations Act, which legally protected workers’ rights to organize unions, restricted businesses ability to interfere with or fire unionizing workers, and enabled compulsory union membership in organizations where unions had been established. Between World War II and the mid-1950s, union density peaked at around one-third of the work force. Union density began a secular decline after 1954, and is quickly approaching levels that prevailed before the passage of the NLRA.

The figure depicts the decline of union density, which is the proportion of workers in unions. This decline in density does not necessarily represent a decline in the absolute number of union jobs, but is more a reflection of a long-term faster growth in non-union jobs. However, after 1980, the absolute number of union jobs began to fall. In 1980, there were approximately 20 million union jobs, whereas there were about 14.5 million in 2013.2 This decline followed several changes, including the deinstitutionalization of regulations that benefitted unions (e.g., “right to work” legislation), employment declines in traditionally unionized sectors (e.g., automotive, utilities), and an increasingly cultural antipathy towards unionism.

Insofar as union membership is concerned, it is clear that the institution of unionism has declined considerably over the past several decades. Union jobs are less prevalent and decreasing in number. There are disagreements about whether or not the decline of unionism is a good or bad thing for workers and society-at-large, but it seems clear that this decline is taking place.


  1. Pre-1973 ata from Alejandro Donado and Klaus Walde (2012) “How Trade Unions Increase Welfare” Economic Journal 112(563): 990 – 1009. Set draws strongly from Richard B. Freeman (1998) “Spurts in Union Growth: Moments and Social Processes” in Michael D. Bordo, Claudia Goldin, and Eugene N. White (eds.) The Defining Moment: The Great Depression and the American Economy in the Twentieth Century University of Chicago Press. Post-1973 numbers from Barry Hirsch and David Macpherson “Union Membership, Coverage, Density, and Employment, Among All Wage and Salary Workers, 1973-2014” Unionstats.com
  2. Donado and Walde (2012) Op. Cit.

Download the Markdown file and raw data

Immigration Boom: Levels Back to 19th Century Levels

Immigration has been a hot button issue in both the United States and over much of Europe. Has immigration risen substantially? We examine the issue below.

How Big Is the Immigrant Population?

Compared to previous decades, the immigrant population is large.  The figure below charts out the ratio of migrants to the overall population since 1850.  It uses Census data compiled by the Migration Policy Institute.

immigrantpopulation

Throughout much of the 19th century, America’s borders were largely open to immigration. America began to restrict immigration at the beginning of the 20th century.  Its immigrant stock fell as immigrants died and were not replaced by new immigration. By the 1970s, immigration reached a low point, after which the country progressively opened its borders. By the 2010s, the country’s immigrant stock is roughly where it was during the 19th century.

The Changing Composition of Immigration

Not only has the number of immigrants risen, but so has teh composition of the immigrant population. The Migration Policy Institute’s Jie Zong and Jeanne Batalova note:

In 2014, Mexican immigrants accounted for approximately 28 percent of the 42.4 million foreign born in the United States, making them by far the largest immigrant group in the country. India, closely trailed by China (including Hong Kong but not Taiwan), and the Philippines were the next largest countries of origin, accounting for about 5 percent each. El Salvador, Vietnam, Cuba, and Korea (3 percent each), as well as the Dominican Republic and Guatemala (2 percent each), rounded out the top ten. Together, immigrants from these ten countries represented close to 60 percent of the U.S. immigrant population in 2014.

The predominance of Latin American and Asian immigration in the late 20th and early 21st centuries starkly contrasts with the trend seen in 1960 when immigrants largely originated from Europe. Italian-born immigrants made up 13 percent of the foreign born in 1960, followed by those born in Germany and Canada (about 10 percent each). In the 1960s no single country accounted for more than 15 percent of the total immigrant population.

So, not only have the raw numbers risen, but their composition has changed. In contrast to fifty years ago, immigrants come mainly from non-white or Hispanic countries, and today’s migrant population has one large group from a common origin (Mexico).

Anxiety over Immigration

A rising tide of non-white and Hispanic immigrants, and a perceived large influx of a particular community, evokes anxiety among those who fear or harbor animosity towards these ethnic groups. These groups are often blamed with damaging the economic fortunes of native Americans, while pushing up crime. The data suggests that there is little basis to arguments linking immigration to crime – immigrants are more law-abiding than natives, and crime has generally been falling during this long growth in the immigration stock. The argument linking immigration to the native-born population’s economic problems is far more complex, and there is little reason to believe that people would be better off economically if immigration were to be cut (I will save that for another entry).

Still, the high level of salience attached to immigration in our policy debates reflects the fact that it is a real and major trend. America is certainly becoming a national of immigrants again.