Immigration Levels are Key to Future Apartment Demand
Will rental apartment production stagnate or grow in the years ahead? The answer to this question is likely to depend on future immigration levels. Recently released data culled from the 2000 Census verify the importance of immigration to the rental apartment market. We now know that the government systematically underestimated immigration levels during the late 1990s—a fact that largely explains why multifamily starts and absorption rates were so strong then, stronger than virtually all experts had predicted.

The Census Bureau’s most recent projections for immigration include alternate scenarios that we can compare against recent experience. These alternate scenarios include a very conservative estimate for immigration, which the Census Bureau calls its low series and a much more aggressive estimate called its high series. The Census Bureau also did a forecast based on numbers in between the two—called, of course, the middle series. Net immigration has been generally running somewhere between the Bureau’s middle and high scenarios, although most recently, it’s been closer to high than middle. 

Immigration Profoundly Important to Multifamily

These different immigration projections translate into powerful differences in NAHB’s  preliminary long-term forecast for multifamily production over the next decade. When the numbers from the middle immigration assumptions are used, they produce a trend of declining production through 2006, followed by a gradual recovery that doesn’t return to current levels until around 2012.  Using numbers from the high immigration series, on the other hand, generates a consistently rising trend that takes us well above current multifamily production levels beginning in 2005.

 New Data Gives Better Answers

In the short run, demand for rental apartments responds to economic variables such as employment and interest rates.  Over the long term, however, demand for housing tends to be driven simply by changes in the number of households.  Although some of that change is due to changes in the tendency of the population to form households, most of it is the result of population changes in particular age brackets.  

The population in various age brackets, in turn, depends on births, deaths, and net immigration (gross immigration minus emigration). The births have already occurred and are recorded, and deaths are relatively easy to predict, so immigration is the real wild card.

The Census Bureau has just released a new data file from the 2000 Census data (called the Public Use Microdata Sample¹) that allows us to isolate exactly the information we need to look closely at households in rental apartments.

NAHB’s analysis of this new, targeted Census Bureau data shows how important immigration is to the multifamily rental market.  One reason is that immigrants tend to arrive in the U.S. when they are between the ages of 15 and 34, and households headed by people younger than age 35 account for such a large share of the multifamily renters (Figure 1).

                            Source: NAHB tabulations of 2000 Census data; U.S. Census Bureau.

Other segments of the housing market are less dependent on younger households. Only about 15% of condo owners in buildings with 5 or more units – and fewer than 12% of single-family detached home owners – are younger than age 35, for example.

Recent Arrivals Are Even More Likely to Rent

We define recent arrivals as those who have been in the U.S. 8 years or less, primarily because those households most clearly distinguish multifamily renters from other segments of the housing market. While recent immigrants account for almost 9% of multifamily renters, they account for less than 3% of multifamily condo owners and single family detached owners, and less than 1% of single family detached owners (Figure 2).²

                    Source: NAHB tabulations of 2000 Census data; U.S. Census Bureau.

Although a little under 9% may not seem like a huge share of the total market, recent immigrants are very important on the margin as a source of new households, and an increase in the demand for rental apartments.

Moreover, their influence in the market is growing. Recent immigrants accounted for a larger share of the multifamily rental market in 2000 than in 1990 (Figure 3).

               Source: NAHB tabulations of decennial Census data; U.S. Census Bureau.

This helps explain why the multifamily rental population has become more racially and ethnically diverse, a trend that is likely to continue. White Hispanics and Asians, for instance, were each about 4% of the multifamily rental market in 1990.  But by 2000, the shares had grown to 7% and 6%, respectively.³

The 2000 Census also revealed that the government had been underestimating net  immigration by about 200,000 households per year during the 1990s. Rather than the average of 870,000 per year that was estimated, the actual net inflow turned out to be more than 1 million per year. This resulted in an underestimate of  the total number of persons in the U.S. – particularly in the age 25-34 age bracket – as well as those under age 15. The immigration inflow helped compensate for the uneven age distribution created when the 1946-1964 baby boom was followed by a baby bust, and largely explains why multifamily production exceeded expectations during the late 1990s. 

Recent evidence suggests that the immigration tide swelled even further after the 2000 Census, and has been running at a rate of about 1.5 million per year, at least through 2001.

What Are the Implications for the Future?

The fact that the age-15-and-younger population turned out to be larger than originally estimated in 2000 has implications for the multifamily rental market and the construction of new apartments going forward. As that cohort ages, it will begin to form new households, many of whom are likely to rent apartments. As a result, demand for rental apartments and multifamily should increase at least moderately over the period from 2006 to 2012.

The strength of the projected growth, however, depends critically on assumptions about immigration. The Census Bureau has not updated its population projections since evidence from the 2000 Census became available. The current Census middle series projection of 800,000 net immigrants per year is below the rates we’ve experienced since 1990. The high series projection of 1.7 million is above recent experience, but only a little above the most recent figures. 

NAHB’s preliminary baseline long-term forecast assumes the middle immigration series, but forecasts for the number of young households and multifamily production increase substantially if the high series is used. 

Under the middle immigration scenario, the number of households with heads younger than age 35 remains roughly stable at 25.4 million in 2003 and 2004, then actually declines in 2005 and 2006 before it begins to grow again, reaching 26.4 million by 2012.

Under the high immigration scenario, the number of households with heads younger than age 35 increases every year over the next decade, with an increase in the rate of growth after 2005 that takes the under-35 households up to 27.8 million by 2012.

Housing starts in buildings with 5 or more apartments have been above 290,000 every year since 1996. Given the current high vacancy rates and slow rates of absorption of new apartments into the market, the consensus is that this rate of production will (despite a generally improving U.S. economy) decline at least somewhat in the short run.

Assuming the Census middle immigration series, the preliminary baseline forecast shows the annual level of 5-plus starts declining, with some fluctuations, until it dips below 250,000. At some point after that, it will begin climbing until it approaches 290,000 again by 2012.

On the other hand, an alternate forecast based on the high immigration assumption  predicts that 5-plus starts will surge to more than 320,000 in 2005 and grow even further, reaching almost 380,000 by 2012 (Figure 4). This underscores the importance of immigration to the multifamily rental market.

                                         Source: NAHB Long-Term Forecast (preliminary).

A Final Thought to Consider

So far, discussions about homeland security have not produced a serious proposal to curtail immigration, but multifamily developers, owners, and managers will want to pay attention to the discussions to see if that remains the case. As we’ve seen, any significant change in the number of immigrants will affect the future growth and health of the entire multifamily market.

--------------------------------------
¹Technically, the data used is the 1% Public Use Microdata Sample, which contains information on individual persons and housing units for a sample of 1% of the U.S. population.  In order to isolate the segment of the market that developers, owners, and managers of typical multifamily properties would be most interested in, we looked at households paying rent to occupy apartments in buildings with five or more units.  We also left out units where meals are included in the rent, as those would include for the most part apartments intended for seniors, which would tend to have characteristics distinct from other multifamily property. As a shorthand, we refer to households who meet these all criteria as “Multifamily Renters”

² In the figure, multifamily renters are compared only to 1) owners of units in 5+ buildings, 2) renters of single family detached homes, and 3) owners of single family detached homes. To simplify the presentation, single family attached units, structures with 2-4 units, units where meals are included in the rent, units occupied without paying cash rent, and mobile homes, houseboats, etc. are not shown.

³ Changes in the way data on race were collected make comparisons between 1990 and 2000 data inexact. The 2000 Census counted a fairly large share of multifamily renters as being of “other or more than one” race. The 1990 Census, however, did not give respondents the option of classifying themselves as belonging to multiple racial categories.