Donate
Independent, objective, nonpartisan research

R 711LHR

Authors

R 711LHR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3741 (45) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(6) { ["_wp_attached_file"]=> string(12) "R_711LHR.pdf" ["wpmf_size"]=> string(6) "951763" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(57651) "Unauthorized Immigrants in California Estimates for Counties July 2011 Laura E. Hill and Hans P. Johnson with research support from David Ezekiel and Joseph M. Hayes Supported with funding from the Silicon Valley Community Foundation http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 2 Summary California has more unauthorized immigrants than any other state, about 2.6 million of the nation’s 11 million; they make up 7 percent of the total California population and 9 percent of the state’s labor force. For decades, unauthorized immigrants have been a part of California: in many industries in the economy and in rural and urban communities. But recent and comprehensive information about t he numbers and location of this population within California —at the county and sub -county level—does not exist. That this wide information gap exists i s doubly surprising given the amount of energy spent and attention paid to this issue by policymakers and the public over those same decades. This report is the first to use a new source of administrative data at the local level to produce comprehensive and systematic sub -state estimates of the unauthorized immigrant population in California . We find that unauthorized immigrants live in every county in the state, primarily but not only in highly agricultural or highly urban areas. As in the country as a whole, unauthorized workers here reside not just in traditional immigrant communities, but have found homes throughout all regions of the state. Contents Summary 2 Tables 4 Figures 5 Introduction 6 Counting California’s Unauthorized Immigrants 7 Obstacles to Counting the Unauthorized 8 Combining New Administrative Data and Residual Method Data 9 Methodology 12 Where in California Do Unauthorized Immigrants Live? 16 Unauthorized Immigrant Population Time Trends 17 Unauthorized Immigrants Live in Zip Codes t hroughout the State 21 Conclusion 28 References 29 About the Authors 30 Acknowledgments 30 A technical appendix to this paper is available on the PPIC website: http://www.ppic.org/content/pubs/other/711LHR_appendix.pdf http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 4 Tables Table 1 Estimates of unauthorized immigrants in California and the United States (millions) 9 Table 2 Stepwise regression estimates 14 Table 3 Estimates of California county unauthorized immigrant populations (2008) 16 Table 4 Estimates of county unauthorized immigrant populations: preferred model; implied by ITIN tax filings; implied by new noncitizens 19 http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 5 Figures Figure 1 Unauthorized immigrants in California and all other states, selected years 7 Figure 2 Tax returns filed with ITINs 11 Figure 3 Correlations between ITIN tax filers and unauthorized immigrant estimates, all states 12 Figure 4a Estimates of unauthorized immigrants in California, by zip code 22 Figure 4b Estimates of unauthorized immigrants in California, percent of population, by zip code 23 Figure 5a Estimates of unauthorized immigrants, Los Angeles County zip codes 24 Figure 5b Estimates of unauthorized immigrants, percent of population, Los Angeles County zip codes 25 Figure 6a Estimates of unauthorized immigrants, San Francisco Bay Area zip codes 26 Figure 6b Estimates of unauthorized immigrants, percent of population, San Francisco Bay Area zip codes 27 http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 6 Introduction The most basic measures of any population group, its size and location, remain elusive when it comes to unauthorized immigrants in California . Although information about the size of the unauthorized immigrant population in the state—about 2.6 million in 2010 1— is easy to find , no current, comprehensive estimates of this population exist for smaller areas such as counties or cities . Nor are they available for any other state. This information gap creates a problem for local, state, and even federal authorities as they try to evaluate this population and create policies affecting them. It is especially problematic in California, where the unauthorized immigrant population is so large and its location so geographically diverse. In this work, we seek to close this knowledge gap at the sub -state level for California . P roducing sub -state estimates of unauthorized immigrants is challenging because immigrants cannot be counted directly. In this work, we take a unique set of administrative data, in this case IRS tax return data for unauthorized immigrants, and then model the data’s relationship to estimates of unauthor ized immigrants for the states using regression analysis. Because the IRS tax data is available at the zip code level, we ca n use the observed relationship between IRS tax data and state populations of unauthorized immigrants to estimate unauthorized immigrant population s for counties and sub-county areas. Our methodology is one common for demographers estimating population gr owth; both the Census Bureau and the California Department of Finance , for example, use administrative data on births and housing to estimate size and change for large population sets such as age and ethnic groups (although not unauthorized immigrants). A further advantage to our method is that IRS data are released annually. Thus, estimates can be both updated and rep licated across other states that have large unauthorized immigrant populations. We first explain our data and methodology in detail , then pr esent our results. 1 Passel and Cohn (2010 ); Hoefer, Rytina, and Baker ( 2010). http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 7 Counting California’s Unauthorized Immigrants The Pew Hispanic Center (PHC), the Department of Homeland Security (DHS), and Warren ( 2011) produce careful estimates of the size of the unauthorized immigrant population at the state and national level . Because there are no national or state level surveys that obtain the documentation status of the foreign -born residing in the United States, these three sources provide our best indirect estimates of the number and distribution of unauthorized immigrants . This population is in the midst of a major shift. After many years of increases, the number of California’s unauthorized immigrants has remained stable or even declined slightly recently . 2 At the same time, the number living in other st ates has increased substantially compared to California (Figure 1). In 1980, approximately half the nation’s unauthorized immigrants lived in the state, but that share had fallen substantially , to about 26 percent, by 2008. FIGURE 1 Unauthorized immigrants in California and all other states, selected years SOURCE S: Passel and Woodward ( 1984); Warren (2011 ). NOTE: All years except 1980 are from Warren (2011 ). There are other signs of change in national settlement patterns: states with the highest rates of growth of unauthorized immigrants in recent years are not the traditional ones : Mississippi, Alabama, and South Carolina (Warren 2011). Further, the greatest numerical gains in recent years have been in Texas, Florida, and North Carolina . More recently, estimates suggest some states are losing unauthorized immigrant populations —including California, which had 250,000 fewer in 2009 than in 2008, according to DHS 2 Passel and Cohn (2011 ); Warren (2011); Department of H omeland Security ( 2011). 1.0 1.5 2.5 3.0 3.0 2.1 2.1 6.0 8.2 8.5 0 1 2 3 4 5 6 7 8 9 1980 1990200020052008 Unauthorized Immigrants (millions) California All other states http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 8 estimates and 100,000 fewer in 2009 than in 2007, according to the PHC. (The PHC estimate d difference is not statistically significant.) Within California, settlement patterns may be changing as well but until now, there was no way of gauging these. Obstacles to Counting the Unauthorized Because the Census and national population surveys place their primary focus on generating full participation, they do not ask foreign -born participants to reveal their immigration status for fear that they will not participate . The Census, the American Community Survey (ACS) , and the Current Population Survey (CPS) do ask respondents who identify themselves as foreign -born to state their country of birth, their date of arrival, and whether or not they are naturalized U.S. citizens . For those not naturalized, however, no further survey questions provide the detail necessary to determine if they are legally resident (either permanently or temporarily) or unauthorized. A few other surveys come closer to providing counts of the unauthorized than do the census, CPS, or ACS, but none can do the job completely. T he National Agricultural Worker Survey ask s foreign -born participants to state their official immigration status, but their respondents are all agricultural workers, and therefore not representative of the full population of unauthorized immigrants . (Although more than half of all agricultural workers are estimated to be unauthorized, only 4 percent of the nation’s total unauthorized population is employed in agriculture , Passel (2009) finds). The Cal ifornia Health Interview Survey asks respondents to indicate if they are citizens or legal permanent residents, but does not differentiate between temporary visa holders and unauthorized immigrants. Th e New Immigrant Survey provides retrospective information on prior legal status of immigrants who eventually gain legal permanent residency , but the s ample size is too small to pinpoint locations and the se unauthorized immigrants may not be representative of all unauthorized immigrants in all ways . 3 Thus, estimates of unauthorized immigrants at the national and state level are produced indirectly, using what is commonly referred to as a residual technique. We rely on three sources for California data on the unauthorize d that use this technique —PHC, DHS, and Warren (2011). Each is computed using slightly different data and slightly different variants on a residual method approach. To count the foreign- born population, t he PHC estimates use Current Population Survey (CPS) data while the DHS and Warren estimates use the American Community Survey ( ACS). Next, each subtracts estimates of the legal foreign - born residents from the counts of the foreign- born in the CPS or ACS . The remaining, or residual, foreign - born comprise th e estimates of unauthorized immigrants . The three estimates vary somewhat in how they determine the legally resident population of the foreign- born, but all three combine administrative counts of legal admissions with standard demographic techniques. 4 (For more detailed descriptions, see Passel and Cohn 2010, Passel 2007, Hoefer, Rytina, and Baker 2011, and Warren 2011.) The estimates computed by these three are generally in close agreement . 3 Hill, Lofstrom, and Hayes ( 2010) demonstrated many similarities. 4 Warren does not estimate the legal population fully each year. Instead, he estimates legal foreign -born arrivals each year and uses demographic techniques to esti mate changes in the legal population from one year to the next. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 9 TABLE 1 Estimates of unauthorized immigrants in California and the United States (millions) 2000 2005 2006 2007 2008 2009 2010 California Warren 2.6 2.9 2.9 2.9 2.9 PHC 2.3 2.7 2.8 2.7 2.6 2.6 DHS 2.5 2.8 2.8 2.8 2.9 2.6 2.6 Total U.S. Warren 8.6 10.9 11.1 11.2 11.2 PHC 8.4 11.1 12.0 11.6 11.1 11.2 DHS 8.5 10.5 11.3 11.8 11.6 10.8 10.8 SOURCES: Warren (2011); Passel and Cohn ( 2010, 2011); DHS (2010 , 2011 ). In California, credible efforts to estimate some sub-state unauthorized populations have been undertaken, but none are as current as our data and none derive from annually updated, independent, administrative data. Pastor and Marcelli in 2004 published estimates of Mexican unauthorized immigrants in California Public Use Microdata Areas (PUMA s) for 1990 and 2000. 5 These were based on two survey s of Mexican immigrants in Los Angeles County, with the results used to determine characteristics of Mexican unauthorized immigrants. These characteristics were in turn used to assign probabilities of unauthorized status among Mexican immigrant s in the pop ulation data. Heer and Passel (1987) demonstrated that estimates similar in type to those of Pastor and Marcelli ( based on survey data that are applied to population data) matched well with estimates derived by interpolating from estimates based on the res idual method. The Census Bureau did estimate counts of unauthorized immigrants for California counties in the 1980s, but to our knowledge, these estimates were not published, with the exception of Los Angeles County (Heer and Passel 1987) , and were later d iscontinued. More recently, Fortuny, Capps, and Passel (2007) published estimates for five metropolitan areas in California (and 25 such areas nationwide) for the years 2000 and 2003 –2004. In addition, P aral and Associates publishes on its web page estimat es of unauthorized immigrants by U.S. congressional districts for 2000 and 2005. They apportion the Passel estimates to congressional districts based on demographic correlates. Combining New Administrative Data and Residual Method Data As described above, residual method estimates for unauthorized immigrants at the state level are in wide agreement and are generally believed to be the best source of information about unauthorized immigrants in the United States. We use these reliable estimates as the basis from which to derive new estimates for regions, counties, and smaller geographic areas within California , combining them with new administrative data to do so . Because the administrative data we use are not in wide use (despite having been collected since 1996), and have never before been used for this purpose, we devote this section to explaining them and their usefulness for our estimates. 5 PUMAs are larger than zip codes and census tracts. In sparsely populated regions, PUMAs can span many counties, but within dense counties, such as Los Angeles, there may be several. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 10 Since 1996, unauthorized immigrants have been permitted to file federal tax returns using a unique identifier , the Individual Taxpayer Identification Number, or ITIN. Immigrants and the native- born with the right to work use social security numbers (SSN) when filing tax returns, but those without work authorization do not have valid SSNs and should use ITINs instead. 6 Th e IRS has made counts of ITIN filers by zip code publicly available for tax years 2000– 2007 (which correspond to calendar years 200 1–2008). As we show, these counts of ITIN filers are a basis from which to estimate unauthorized immigrants in counties and s ub-count areas in California. Even if they have worked in the United States without proper authorization, unauthorized immigrants are nevertheless required under federal law to file tax returns . Some e stimates suggest that about half do so ( Immigration Policy Center 2011 ; Pastor et al . 2010; Hinojosa -Ojeda 2010), but others show the share is much higher: a recent PPIC report found that over 8 0 percent of unauthorized immigrants reported having filed federal income taxes in the year prior to earning legal permanent residence (Hill, Lofstrom, and Hayes 2010). The Social Security Administration ’s chief actuary estimated that about 75 percent of unauthorized immigrants have payroll taxes withheld (Porter 2005). Unautho rized immigrants have many incentives to file tax returns . First, some who have had taxes withheld by their employers throughout the year would be eligible for tax refunds and might use an ITIN to claim that money . Second, if an unauthorized immigrant ulti mately does become a legal permanent resident and receives an SSN, he or she can link any social security earnings withheld under the ITIN to earnings under the new SSN and have them counted toward later benefits. Third, even unauthorized immigrants who have not had taxes withheld (being self -employed or paid in cash, for example) might file tax returns to establish a positive paper trail for the future: should comprehensive immigration reform ever become a reality, a clear record of employment and tax payments are factors likely to increase an unauthorized immigrant’s chances of attaining legal status . T he majority of ITIN users are unauthorized immigrants, as we explain below. If a former ITIN filer is ever granted the legal ability to work, he or she shou ld begin using his or her assigned SSN immediately. Further, a nyone who is authorized to work in the United States, such as those on temporary work visas (e.g. H -1B or foreign students with work authorization s) are required to apply for and file federal ta xes with an SSN . By December 2008, the IRS had issued more than 14 million ITINs. Not all of these ITINs are used on tax returns filed domestically; s ome are filed from abroad. Many may have been retired after an immigrant was issued a valid SSN. Some may never have been used to file taxes (they may have been obtained to open a bank account, for example). Some may no longer be in use because the ITIN holder no longer files taxes, either b y choice or because he or she no longer is required to do so ( having insufficient income or no longer liv ing in United States). The number of tax returns filed with ITINs and that use U.S. addresses has increased dramatically since 1996. 7 The years for which we have data —2001 to 2008—reveal that many tax filers in California were early users of the ITIN, making up 40 percent of ITIN filers nationally in 2001 , although this fell to 30 percent of filers in by 2008 (Figure 2). Other states have seen similar but smaller changes in their share of ITIN tax filers. 6 ITINs are not a valid proof of employment eligibility; ITIN tax filers could not receive federal stimulus tax rebates, nor ar e they eligible for the Ea rned Income Tax Credit (EITC), which requires a filer, spouse, and child to have valid SSNs. 7 Anomalous increases in ITIN filings were investigated and in some cases, those records were deleted from the totals reported here. Texas and Georgia both had single -year increases in a few zip codes that were larger than the entire zip code p opulation, so we kept the growth in ITIN numbers for those f our zip codes at the level recorded in the previous year. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 11 For example, Ill inois had 4.7 percent of all filers in 2001 and 6.1 percent in 2008. Like California, ITIN tax filers in New York also declined in share of ITIN filers (from 6.5 percent to 5.2 percent ). North Carolina ’s share rose (from 1.8 percent to 3.7 percent over the same period). These changes are largely consistent with the residual estimates of unauthorized immigrants in these states. For example, California’s share of the nation’s unauthorized population fell from 43 percent in 2000 to 35 percent in 2008, according to the Warren estimates. FIGURE 2 Tax returns filed with ITINs SOURCE: Authors’ calculations of ITIN numbers . NOTE: Beginning in 2006, dependents with ITINs were also recorded. In California, nearly 6 percent of tax filers used an ITIN in 2008 (929,000) , up from 2 percent in 200 1. These include primary filers , spouses, or dependents . Although not all ITIN filers were unauthorized, the number of ITIN filers was 36 percent of the number of estimated unauthorized immigrants. A lthough ITIN tax filing data may serve as a good proxy for unauthorized immigrants, they can not provide a precise count , for two reasons . First, not all unauthorized immigrants file taxes, and among those that do, not all use the ITIN . Some may file instead using a false, fraudulent SSN, or an SSN issued many years ago that did not permit work 8 or one that no longer does . Second, not all ITIN filers are unauthorized immigrants . H owever, once we exclude tax filers from abroad, the vast majority of ITIN filers do appear to be unautho rized. When we examine ITIN tax filers with U.S. filing addresses, we find that 90 percent in 2008 include wages. Only unauthorized workers would file tax retu rns with wages and an ITIN ; authorized workers with wages would file tax returns using SS Ns. Therefore, it is safe to conclude that the vast majority of ITIN fil ers are unauthorized workers . 8 These “no work” SSNs are no longer issued. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 2001 2002200320042005200620072008 Taxes filed with ITINs (millions) Total California Florida New York Texas http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 12 Methodology Using ITIN data to estimate the distribution of unauthorized immigrants within California (or any state) depends on a strong relationship between the ITIN numbers and existing estimates of unauthorized immigrants. ITIN numbers will not include all unauthorized immigrants, nor will they include only unauthorized immigrants . But to use ITINs as a basis from which to scale local area estimates to match state total estimates we need the correlation between ITIN counts and unauthorized estimates to b e high. And indeed, we find ITINs are highly correlated with independently derived state estimat es of unauthorized immigrants. We calculate d correlations for each year of available ITIN data with the Warren and PHC estimates (Figure 3 ). Each point in the figure represents that year’s correlations for the 50 states and the District of Columbia. FIGURE 3 Correlations between ITIN tax filers and unauthorized immigrant estimates, all states SOURCES: Authors’ calculations from ITIN counts ; Warren (2011); Passel and Cohn ( 2011, 2010). NOTE: 2008 ITIN correlations for Passel and Cohn use 2009 data; DHS estimates are available only for the 10 states with the largest unauthorized populations . The Warren estimates have a correlation value of nearly 0.98 with the ITIN tax filing records (see Appendix Table A 1 for the full set of Warren estimates) . The PHC estimates use ranges for states with small populations, which may partially explain their lower correlation , which is still very high: in excess of 0.955 for all years . In the PHC estimates, those of 32 states are derived using multiple years of data because of the very small CPS data sample of likely unauthorized immigrants in those states—fewer than 50 in each year ( Passel and Cohn, 2011) . Altogether, these extremely high correlations give us great confidence that ITIN filings are an excellent indicator of the unauthorized immigrant population. We also examined the relationship between the estimates of unauthorized immigrants and ITIN data filed with wages attached (W -2s) , ITIN taxing filings prepared by a paid tax preparer, and ITINs not filed as a n on resident. Each of these has the potential to be more highly correlated with unauthorized immigrants than the overall ITIN filing rates. ITI N returns filed with wages are almost definitely unauthorized immigrant tax filers (all other filers with the legal right to work should have SSNs), b ut not all ITIN unauthorized tax filers 0.9450.95 0.955 0.96 0.965 0.97 0.975 0.98 0.985 0.99 2001 2002200320042005200620072008 Correlations Warren Pew Hispanic Center http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 13 will have wages attached (e.g. those who file but are paid cash or are self -employed ). The vast majority of ITIN filers use paid preparers , and do so at much higher rate than tax filers in general . This may be because unauthorized immigrant tax filers may prefer to pay for help in filing correctly if their motivation is to document earnings and tax paying ; or it may simply be that unauthorized tax filers who use paid preparers are m ore likely to learn about ITINs than if they do not . 9 Unauthorized immigrants should also be more likely to file as residents (rather tha n using the nonresident 1040NR form) because nonresidents can not claim child tax credits and because filing as a nonresident might do less to establish one’s intent to ultimately legalize. 10 However, we found n one of these other measures of ITIN tax fil ers were as consistently highly correlated as the correlations with the overall ITIN tax filings for the states. Given the high correlation between ITIN filers and estimates of the unauthorized for the 50 states and D .C., we could use the simple ratio of unauthorized immigrants to ITIN filers as the factor to estimate local populations of unauthorized immigrants . However, we know this ratio varies across states and time , and so suspect that it might also vary within the state. We use regression analyses to account for this cross-state and intra- state variation . Differences in the rate of ITIN usage by a state’s unauthorized population may be related to variation s in the characteristics of employment and earnings among unauthorized immigrants, their demograp hic characteristics , or infrastructure available to support immigrant tax filing, among others . Using the Warren estimates, ITIN data, the ACS (for the 2006 and 2008 models) , and the 2000 c ensus (for the 2001 model ), we use weighted least squares (WLS) to estimate the following regression model of the ITIN coverage rate for each state (s) and each year ( t), with proportionate weight applied to the estimated size of the undocumented population . We restrict the population to foreign -born residents : ��������� ������������ ����������������� ���� = ������α + ������β+ ������γ+������ In the equation, X represents a matrix of demographic characteristics of the immigrant population in each state and each year . It includes age, proportion Latino, and proportion born in Central America; w e do not include the proportion born in Mexico because that is so closely correlated with the proportion s that are Latino and Central America n-born . Employment characteristics for the immigrant population are represented by the mat rix W and include the share employed in construction, the share employed in restaurants, proportion self - employed, and proportion not in the labor force. Tax filing characteristics are represented by the matrix Z and include filing as married, being a new tax filer, and filing using a paid preparer. Because we have only 51 observations ( 50 states and D.C. ), we restricted our possible covariates to just a few (Table 2). We want these relationships to be able to vary, first because the increase in ITIN usage was so great during th e interval we studie d (Figure 2), and second, because we do not fully understand why some states seemed to have a higher percentage of unauthorized immigrants filing income taxes with ITINs . We therefore allowed the variables that en tered into this equation to change across time . We used a backward elimination stepwise regression method, removing the least significant variable from the model one at a time until all variables met a predetermined threshold of significance. In this case our threshold wa s a p-value of 0 .10. Regressions were run separately for each year. 9 Filers claiming the EITC also use paid preparers at a very high rate. In California, 76 percent of EITC filers used paid preparers in 2006 (Danielson 2010). 10 1040NR tax forms can be filed from U.S. addresses by those who are only temporarily residing in the U.S. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 14 In developing our model, we considered different population bases to use to estimate the ratio of ITIN tax filers to Warren estimates , and settled on the foreign -born as the base population of each state . 11 We discarded some of the variables that were determined to be significant based on the state- level models because the range of values for those variables for counties in California was far outside the range of values for states. For example, the share of the foreign -born population working in agriculture ranges from l percent to 25 percent in the 50 state s and D .C., but in California counties, the high end is much greater: Tulare County has 61 percent of its foreign -bo rn labor force working in agriculture, and many other California counties were above 40 percent. We considered models in which we predicted the ITIN numbers, with the Warren estimates on the right hand side of the equation. We also predicted the logged IT IN numbers. Both of these models resulted in state -level computed estimates that had a poorer fit than the ratio models—that is, our predictions for states were not as close to the actual Warren estimates as in our final model . Further , we estimated our models using the PHC estimates as robustness check. Because correlations are lower (Figure 3) and the estimates are not available for all year s, we prefer the Warren estimates . (The models estimated with the PHC data are available on request.) Our final mo del varied for each year. We report those for 2001, 2006, and 2008 below; Table 2 reports all of the variables that we allowed to enter into the stepwise regression . As noted above, only those variables which were estimated with a p value of 0.10 were ultimately included in the regression . We have the most confidence in the county estimates for 2008 . TABLE 2 Stepwise regression e stimates 2001 2006 2008 R squared 0.701 0.593 0.618 hettest 0.012 0.501 0.692 Coeff p-value Coeff p-value Coeff p-value Age 0 –17 Age 35–54 0.215721 0.038 – Age 55+ 0.186604 0.006 Proportion Latino Born in Central America -0.1601 0.000 -0.41705 0.000 -0.5378 0.000 Construction Restaurants Self-employed Not in the labor force Filed taxes as married -0.11164 0.000 -0.39599 0.000 -0.48947 0.000 New tax filer Filed using paid preparer 0.065578 0.064 Constant 0.015552 0.718 0.489259 0.000 0.598122 0.000 SOURCES: Authors’ calculations using ITIN, ACS, 2000 census, and Warren and Associates data . 11 We also considered models using re cently arrived foreign -born noncitizens as the base population . Those models had so much more variation within California counties than across the 50 states and strained the ability of regression to make useful estimates. Expanding the sample more widely to include the entire population resulted in a poorer fitting model as well. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 15 In each of the years 2001, 2006, and 2008, the share of the foreign -born population from Central America and the share of ITIN tax filers that filed tax returns as married was very important in scaling the ITIN returns to m atch the state -level estimates of unauthorized immigrants . Also important for 2001 was the share of the foreign -born population between ages 35 and 54, and 55 or o lder, as was the share of ITIN tax filers who used a paid tax preparer. Applying these coeff icient estimates for the states to ITIN tax -filer and ACS data for the counties, we computed a county - (or regional -) level count of unauthorized immigrants. 12 These were then totaled and scaled to match the estimate of unauthorized immigrants for the state in that year . Our final step was to scale these local estimates back down to the zip code level, using the distribution of ITIN filings filed by zip code within that county . (We could not use our model and ACS data in the same way because ACS data are not available for that small geography.) 12 We totaled zip codes to the county or regional level. Many zip codes span two (or more) counties, and are allocated to counti es based on 2000 census block populations ( Kneebone 2008). http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 16 Where in California Do Unauthoriz ed Immigrants Live? For 2008, we find unauthorized immigrants residing in all counties (or county groups) throughout the state. 13 Unauthorized immigrant s are found in major urban areas, agricultural regions, and places in between. According to our estimates, their raw numbers range from just over 1,000 in the Del Norte/Siskiyou/Modoc/ Lasse n county grouping to just under 1 million in Los Angeles County (Table 3). In general, unauthorized immigrants make up small but notable shares of county populations. In only four counties or county groupings do they make up more than 10 percent of the total population. In 22 counties, mostly rural and mountainous, but also including Sacramento, unauthorized immigrants make up less than 5 percent of the population. Not surprisingly, the most populous counties have the largest populations of unauthorized immigrants . TABLE 3 E stimates of California county unauthorized i mmigrant populations (2008) County Total population (2008 ACS estimate) Unauthorized immigrants Population estimate % of county total population Alameda 1,475,000 124,000 8.4% Amador, Calaveras, Tuolomne, Mariposa, Alpine, Mono, Inyo 191,000 2,500 1.4% Butte 220,000 4,000 1.8% Colusa, Glenn, Tehema, Trinity 124,000 10,000 8.3% Contra Costa 1,029,000 79,000 7.7% Del Norte, Siskiyou, Modoc, Lassen 118,000 1,000 1.0% El Dorado 176,000 4,000 2.2% Fresno 909,000 49,000 5.3% Humboldt 129,000 2,000 1.6% Imperial 164,000 21,000 12.8% Kern 801,000 46,000 5.7% Kings 150,000 9,000 5.8% Los Angeles 9,860,000 916,000 9.3% Madera 149,000 12,000 7.7% Marin 249,000 14,000 5.6% Mendocino, Lake 151,000 8,000 5.0% Merced 246,000 22,000 9.1% Monterey, San Benito 463,000 62,000 13.5% Napa 134,000 16,000 12.0% 13 Because the ACS is a sample, not all counties have large enough populations to be reported separately in it. Because of these ACS sample size restrictions, we report unauthorized immigrant estimates for 34 counties and 7 county groups rather than 58 counties for both 2001 and 2008. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 17 County Total population (2008 ACS estimate) Unauthorized immigrants Population estimate % of county total population Orange 3,010,000 289,000 9.6% Placer 342,000 8,000 2.3% Plumas, Sierra, Nevada 120,000 2,000 1.5% Riverside 2,101,000 146,000 7.0% Sacramento 1,394,000 65,000 4.6% San Bernardino 2,015,000 150,000 7.5% San Diego 3,002,000 198,000 6.6% San Francisco 809,000 30,000 3.7% San Joaquin 673,000 54,000 8.0% San Luis Obispo 266,000 9,000 3.5% San Mateo 712,000 55,000 7.8% Santa Barbara 405,000 37,000 9.0% Santa Clara 1,764,000 180,000 10.2% Santa Cruz 253,000 21,000 8.2% Shasta 180,000 1,000 0.6% Solano 407,000 24,000 6.0% Sonoma 467,000 41,000 8.8% Stanislaus 511,000 39,000 7.6% Sutter, Yuba 165,000 9,000 5.6% Tulare 426,000 29,000 6.8% Ventura 798,000 74,000 9.3% Yolo 198,000 12,000 6.2% Total 36,756,000 2,876,000 7.8% SOURCE S: Authors’ calculations ; ACS . Unauthorized Immigrant Population Time Trends In this section, we present our estimates for 2001 and 2008 together. We expect ed that our method and model s would be more reliable for years when ITIN numbers were more commonly used than the years when ITINs were new and less likely to be used by unauthorized immigrants. However, the fit for our model that predicted the ratio of ITIN filings to the Warren unauthorized estimates was actually slightly better in 2001 than 2008 (Table 2), despite the fact that, as Figure 2 illustrated , there was a dramatic uptick in use of ITINs among tax filers. We also found that from 2001 to 2008 the number of zip codes with ITIN tax filers increased. Taken individually, the estimates for the single years seem reasonable. Although our 2001 model fits well, we are still cautious about our results from years before 2008 mainly because a smaller share of unauthorized immigrants was filing ITIN returns in the earlier years; when w e examine the change from 2001 to 2008, we have less certainty about the prior years . We find that for many of the small county and small county groupings, the growth in unauthorized immigrants that is implied from our estimates is perhaps TABLE 3 (continued) http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 18 implausible. Therefore, the 2001 estimates we present are only for those counties with 2008 populations of 200,000 or greater and those in which our 200 1 estimate of unauthorized immigrants was greater than 10,000. We offer two benchmarks , for comparison only . The first is the simple distribution of the estimated number of unauthorized immigrants using the ITIN counts for counties from the administrative data. Comparing our model estimates to the ITIN results gives a sense of how our model may be an improvement over simply scaling the administrative tax data . The second benchmark is the distribution of the estimate of unauthorized immig rants using the distribution of the state’s new noncitizens (arrived within the last 20 years ) to counties. This is one way to allocate the reputable state estimates to sub -state areas (but not a method employed by any of those who compute those residual methods) . Our method, because its underlying data are available every year, for all zip codes nationwide, and because it does not rely on any other allocation or estimation (with the exception of state -level estimates) , is the best methodology available gi ven the current data constraints . http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 19 TABLE 4 E stimate s of county u nauthorized immigrant populations : preferred model ; implied by ITIN tax filings; implied by new non citizen s* 2001 2008 Change 2001 to 2008 County County population (2008 ACS) Model ITIN filers New non-citizens Model ITIN filers New non-citizens Model ITIN filers New non-citizens Alameda 1,475,000 163,000 144,000 121,000 124,000 116,000 116,000 -39,000 -28,000 -5,000 Amador, Caleveras, Tuolmne, Mariposa, Alpine, Mono, Inyo 191,000 2,500 3,000 2,000 Butte 220,000 4,000 4,000 6,000 Colusa, Glenn, Tehema, Trinity 124,000 10,000 10,000 6,000 Contra Costa 1,029,000 63,000 63,000 50,000 79,000 71,000 71,000 16,000 8,000 21,000 Del Norte, Siskiyou, Modoc, Lassen 118,000 1,000 1,000 1,500 El Dorado 176,000 4,000 4,000 4,000 Fresno 909,000 30,000 25,000 63,000 49,000 51,000 72,000 19,000 26,000 9,000 Humboldt 129,000 2,000 2,500 1,000 Imperial 164,000 21,000 18,000 14,000 Kern 801,000 21,000 18,000 36,000 46,000 46,000 53,000 25,000 28,000 17,000 Kings 150,000 9,000 9,000 12,000 Los Angeles 9,860,000 924,000 948,000 1,069,000 916,000 894,000 987,000 -8,000 -54,000 -82,000 Madera 149,000 12,000 12,000 12,000 Marin 249,000 16,000 19,000 12,000 14,000 14,000 12,000 -2,000 -5,000 0 Mendocino, Lake 151,000 8,000 8,000 5,000 Merced 246,000 15,000 12,000 16,000 22,000 24,000 24,000 7,000 12,000 8,000 Monterey, San Benito 463,000 39,000 37,000 43,000 62,000 73,000 48,000 23,000 36,000 5,000 Napa 134,000 16,000 15,000 12,000 Orange 3,010,000 349,000 387,000 273,000 289,000 323,000 267,000 -60,000 -64,000 -6,000 Placer 342,000 8,000 7,000 10,000 Plumas, Sierra, Nevada 120,000 2,000 2,000 500 http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 20 2001 2008 Change 2001 to 2008 County County population (2008 ACS) Model ITIN filers New non-citizens Model ITIN filers New non-citizens Model ITIN filers New non-citizens Riverside 2,101,000 78,000 75,000 85,000 146,000 145,000 145,000 68,000 70,000 60,000 Sacramento 1,394,000 42,000 36,000 60,000 65,000 64,000 80,000 23,000 28,000 20,000 San Bernardino 2,015,000 100,000 99,000 95,000 150,000 141,000 125,000 50,000 42,000 30,000 San Diego 3,002,000 189,000 165,000 175,000 198,000 184,000 177,000 9,000 19,000 2,000 San Francisco 809,000 42,000 51,000 64,000 30,000 33,000 65,000 -12,000 -18,000 1,000 San Joaquin 673,000 31,000 27,000 35,000 54,000 51,000 43,000 23,000 24,000 8,000 San Luis Obispo 266,000 9,000 10,000 9,000 San Mateo 712,000 64,000 71,000 60,000 55,000 55,000 65,000 -9,000 -16,000 5,000 Santa Barbara 405,000 37,000 36,000 29,000 37,000 39,000 35,000 0 3,000 6,000 Santa Clara 1,764,000 241,000 246,000 182,000 180,000 185,000 190,000 -61,000 -61,000 8,000 Santa Cruz 253,000 15,000 17,000 16,000 21,000 24,000 17,000 6,000 7,000 1,000 Shasta 180,000 1,000 1,000 2,000 Solano 407,000 16,000 15,000 16,000 24,000 23,000 22,000 8,000 8,000 6,000 Sonoma 467,000 42,000 43,000 22,000 41,000 43,000 26,000 -1,000 0 4,000 Stanislaus 511,000 23,000 22,000 24,000 39,000 38,000 28,000 16,000 16,000 4,000 Sutter, Yuba 165,000 9,000 9,000 10,000 Tulare 426,000 33,000 33,000 29,000 29,000 32,000 33,000 -4,000 -1,000 4,000 Ventura 798,000 48,000 49,000 45,000 74,000 83,000 54,000 26,000 34,000 9,000 Yolo 198,000 12,000 12,000 12,000 Total 36,757,000 2,711,000 2,711,000 2,711,000 2,876,000 2,876,000 2,876,000 165,000 165,000 165,000 SOURCE S: Authors’ calculations ; 2008 ACS . * New noncitizens are non- naturalized foreign-born who arrived in the previous 20 years (ACS 2008). TABLE 4 (continued) http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 21 According to our estimates, the population of unauthorized immigrants in Los Angeles County was 924,000 in 2001 and declined very slightly to 916,000 by 2008. Two other sources have also estimated this population for similar years; the Los Angeles Family and Neighborhood Survey of 2000 f ound that there we re about 664,000 unauthorized adults in Los Angeles County. Using the residual method, Fortuny, Capps, and Passel (2007) estimate d about 937,000 in the same year, 2000 . 14 Our estimates are much closer to those that use the residual approach . Similarly, c hanges over time in large counties appear to be smaller proportionally, according to our model , than unauthorized immigrant population changes in the smaller counties . Unauthorized Immigrants L ive in Zip Codes t hroughout the State Within counties, we are able to match, approximate ly, the unauthorized immigrant population to the zip codes of residence. As we described above, however, we cannot use our model to directly estimate unauthorized immigrant residents in zip codes . Instead, we take our county -level estimates (derived from our models ), and allocate the unauthorized immigrants based on the distribution of ITIN tax return filers by zip code within the county . There are three potential problems with this approach. First, counts of fewer than 10 ITIN tax filers at the zip code level were suppressed by the IRS , so county totals of unauthorized immigrants cannot be allocated to that zip code. However, the number of zip codes with 10 or more ITIN tax filers has risen rapidly in our data years. In 2001, 54 percent of zip codes with any tax filers had ITIN tax filers; by 2008, th at share had risen over two thirds, 67 percent. Second, some zip codes are actually points, such as post office boxes, or office buildings . These are not mapped, but the data from them are included in our county estimates (Tables 3 and 4 ). Third, because tax filers may use a work address or an address other than a residence, we find that in a few zip codes we predict higher numbers of unauthorized residents than there are total residents . Th ese are few: in 2008, we found nine , defined as zip codes where the total population was fewer than 1, 000 and the per centage of unauthorized residents was greater than 35 percent. In our maps that display the percentage of zip code residents that are unaut horized immigrants, we do not show levels over 15 percent , and so do not expect that these nine zip codes dramatically alter the visual presentation of our results. O ur methods clearly cannot predict unauthorized immigrants residing in the state’s zip codes with exact precision . For that reason, we present our zip code results in ranges, rather in specific number or tabular form . In addition, we do not separate zip codes with zero unauthorized immigrants from zip codes with just a few unauthorized immigrants because of the IRS data suppression issue. Maps of the state by zip code reveal unauthorized immigrants residing in some very highly concentrated pockets throughout the state, but also located in some places of relative isolation . Throughout the state, we find zip codes with more than 5 ,000 unauthorized immigrant residents well outside highly urbanized areas (Figure 4 a). When we consider the unauthorized as a percentage of the populat ion, we find many zip codes where 15 percent of the population is unauthorized acros s even more diffuse and diverse geographies (Figure 4b). Maps for Los Angeles County (Figures 5a and 5b) and for the San Francisco Bay area (Figures 6a and 6b) are provided to illustrate the patterns that emerge from estimating sub -country distributions . All maps reflect 2008 data. 14 Fortuny et al. (2007) also provide estimates for Los Angeles County in 2003-04 (1,000,000), Orange County PMSA (245,000 in 2000), and Riverside/San Bernardino PMSA (175,000 in 2000). http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 22 FIGURE 4A Estimates of unauthorized immigrants in California , by zip c ode SOURCE: Authors’ calculations using ITIN and ACS data. NOTE : Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 23 FIGURE 4B Estimates of u nauthorized immigrants in California, percent of population, by zip code SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 24 FIGURE 5A Estimates of unauthorized immigrants, Los Angeles County zip codes SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 25 FIGURE 5B Estimates of u nauthorized immigrants , percent of population, Los Angeles County zip codes SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 26 FIGURE 6A Estimates of unauthorized immigrants , San Francisco Bay Area zip codes SOURCE: Authors’ calculations using ITIN and ACS dat a. NOTE: Areas in white indicate no population . http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 27 FIGURE 6B Estimates of u nauthorized immigrants , percent of population, San Francisco Bay Area zip codes SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp 28 Conclusion We have developed comprehensive sub-state estimates of unauthorized immigrants for California —perhaps the most important state to understand when considering i mmigration issue s. Our estimates are based on administrative data— income tax returns by unauthorized immigrant s—available for local areas. Prior to the ir availability, the best es timates about where in the state unauthorized im migrants reside were limited to larger levels of geography and are now either outdated or avail able only for subsets of th is population. As with any estimates of unauthorized immigrants, these numbers are subject to uncertainty . However, t he administrative data we rely on is highly correlated with independently developed residual estimates of state unauthorized immigrant populations . We take further comfort in our results for Los Angeles County, which are consistent with other estimates derived from the residual method. We expect that as the percentage of unauthorized immigrants using ITIN numbers on their tax returns increases across the country , our method could be used to compute similar sub -state estimates in other locations. Currently, it is reasonable to attempt to do so with states with large populations of unauthorized immigrants who appear to be using ITINs in large number s, such as Texas, New York, Illinois, and Arizona.     http://www.ppic.org/main/home.asp 29 References Danielson,  Caroline.  2010.  “The  Earned  Income  Tax  Credit  in  California.”  Just  the  Facts.  Public  Policy  Institute  of  California.   Available  at www.ppic.org/main/publication.asp?i=925 .  Fortuny,  Karina,  Randy  Capps,  and  Jeffrey  S.  Passel.  2007.  “The  Characteristics  of  Unauthorized  Immigrants  in   California,  Los  Angeles  County,  and  the  United  States.”  Washington  DC:  Urban  Institute.   Goldman,  Dana P.,  James  P.  Sm ith ,  and  Neeraj  Sood.  2006.  “Immigrants  and the  Cost  of  Medical  Care.”  Health  Affairs  25  (6).   Heer,  David  M.,  and  Jeffrey  S.  Passel.  1987.  “Comparison  of Two  Methods  for  Estimating  the Number  of  Undocumented   Mexican  Adults  in  Los  Angeles  County.”  International  Migration  Review  21  (4).    Hill,  Laura  E.,  Magnus  Lofstrom , and  Jo seph M.  Hayes.  2010.  Immigrant  Legalization:  Assessing  the  Labor  Market  Effects.  San  Francisco:  Public Policy Institute  of  California.  Available  at www.ppic.org/main/publication.asp?i=869.   Hinojosa ‐Ojeda, Raul.  2010.  “Raising  the  Floor  for  American  Workers:  The Economic  Benefits  of  Comprehensive   Immigration  Reform.”  Washington  DC:  Center  for  American  Progress  and Immigration  Policy  Center.   Hoefer,  Michael,  Nancy  Rytina,  and  Bryan  C.  Bak er.  2011.  “Estimates  of  the  Unauthorized  Immigrant  Population   Residing  in  the  United  States:  January  2010.”  Population  Estimates,  Office of  Immigration  Statistics,  Department  of   Homeland  Security.    Immigration  Policy  Center.  2011.  “Unauthorized  Immigrants  Pay  Taxes,  Too:  Estimates  of  the  State  and  Local  Taxes  Paid   by  Unauthorized  Immi grant Households.”   Internal  Revenue  Service. SPEC  Tax  Data.  Available  through  Brookings  EITC  interactive  ( www.brookings.edu ).  Kneebone,  Elizabeth.  2008.  “Bridging  the  Gap:  Refundable  Tax  Credits  in  Metropolitan  and  Rural  America.”  Earned   Income  Tax  Credit  Series,  Metropolitan  Policy  Program  at  Brookings.  Washington,  DC:  Brookings  Institute.   Passel,  Jeffrey.  2007.  “Unauthorized  Immigrants  in  the  Unit ed  Stat es: Estimates,  Methods,  and  Characteristics.”  OECD   Social,  Employment  and  Migration  Working  Paper.    Passel,  Jeffrey  S.,  and  Karen  A.  Woodrow.  1984.  “Geographic  Distribution  of  Undocumented  Immigrants:  Estimates  of   Undocumented  Aliens  Counted  in  the  1980  Census  by  State.”  International  Migration  Review  18  (3).    Passel,  Jeffrey  S.,  and  D’Vera  Cohn.  2011.  “Unauthorized  Immigrant  Popu lation:  Na tional and  State  Trends,  2010.”   Washington  DC:  Pew  Hispanic  Center.  Passel,  Jeffrey  S.,  and  D’Vera  Cohn.  2010.  “U.S.  Unauthorized  Immigration  Flows  Are  Down  Sharply  since  Mid ‐Decade.”   Washington  DC:  Pew  Hispanic  Center.  Passel,  Jeffrey  S.,  and  D’Vera  Cohn.  2009.  “A  Portrait  of  Unauthorized  Immigrants  in  the  United  States.”  Washington,   DC:  Pew  Hispanic  Center.  Pastor,  Manuel ,   Justin  Scoggins,  Jennifer  Tran,  and  Rhonda  Ortiz.  2010.  “The  Economic  Benefits  of  Immigrant   Authorization  in  California.”  Los  Angeles:  Center  for  the  Study  of Immigrant  Integration,  University  of  Southern   California.   Pastor  Jr.,  Manuel,  and  Enrico  A.  Marcelli.  2004.  “Somewhere  Over the  Rainbow?  African  Americans,  Unauthorized  Mexican  Immigrati o n,  and  Coalition  Building.”  In  The  Impact  of  Immigration  on  African  Americans,  ed.  Steven   Schulman  (Piscataway,  NJ:  Transaction  Publishers).   Porter,  Eduardo.  2005. “Illegal  Immigrants  Are  Bolstering  Social  Security  with  Billions.”  New  York  Times,  April  5.    Rob  Paral  and  Associates.  “Undocumented  Immigrants  in  Congressional  Districts.”  Available  at   www.robparal.com/MapPage.html?map=14&type=G . Accessed  May 5,  2011  (Google  Earth  plug‐in  required).   Warren,  Robert.  2011.  “Annual  Estimates  of  the  Unauthorized  I mmigrant  Population  in  the  United  States,    by  State:  1990  to 2008.”  Working  paper,  Public  Policy Institute  of  California.  Available  at  http://www.ppic.org/main /publication.asp?i=992.  http://www.ppic.org/main/home.asp Title 30 About the Author s Laura E. Hill is a policy fellow at the Public Policy Institute of California. Her research interests include immigrants, immigration, race and ethnicity, and youth. She has been a research associate at The SPHERE Institute and a National Institute of Aging postdoctoral fellow. Sh e holds a Ph.D. in demography from the University of California, Berkeley. Hans Johnson is a senior policy fellow at the Public Policy Institute of California. His research focuses on the dynamics of population change in California and policy implications of the state’s changing demography. At PPIC, he has conducted research on international and domestic migration, population projections, housing, and higher education. Before joining PPIC as a research fellow, he was senior demographer at the California Re search Bureau, where he conducted research on population issues for the state legislature and the governor’s office. He has also worked as a demographer at the California Department of Finance, specializing in population projections. He holds a Ph.D. in de mography from the University of California, Berkeley. Acknowledgments The authors wish to thank Robert Warren and Elizabeth Kneebone and for their help with the data. Technical reviews of the method and paper by Jeffrey Passel and Manuel Pastor are much appreciated. We also thank Helen Lee, Kim Belshé, Robert Gleeson, and Abby Cook for multiple reviews and conversations, and Richard Greene for his skillful wordsmithing. PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors John E. Bryson, Chair Retired Chairman and CEO Edison International Mark Baldassare President and CEO Public Policy Institute of California Ruben Barrales President and CEO San Diego Regional Chamber of Commerce María Blanco Vice President, Civic Engagement California Community Foundation Gary K. Hart Former State Senator and Secretary of Education State of California Robert M. Hertzberg Partner Mayer Brown LLP Walter B. Hewlett Director Center for Computer Assisted Research in the Humanities Donna Lucas Chief Executive Officer Lucas Public Affairs David Mas Masumoto Author and farmer Steven A. Merksamer Senior Partner Nielsen, Merksamer, Parrinello, Gross & Leoni, LLP Constance L. Rice Co- Director The Advancement Project Thomas C. Sutton Retired Chairman and CEO Pacific Life Insurance Company http:// The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute’s goal is to rais e public awareness and to give elected representatives and other decisionmakers a more informed basis for developing policies and programs. The institute’s research focuses on the underlying forces shaping California’s future, cutting across a wide range of public poli cy concerns, including economic development, education, environment and resources, governance, population, public finance, and social and health policy. PPIC is a private operating foundation. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. PPIC was established in 1994 with an endowment from William R. Hewlett. Mark Baldassare is President and Chief Executive Officer of PPIC. John E. Bryson is Chair of the Board of Directors. Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notic e is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. © 2011 Public Policy Institute of California All rights reserved. San Francisco, CA PUBLIC POLICY INSTITUTE OF CALIFORNIA 500 Washington Street, Suite 600 San Francisco, California 94111 phone: 415.291.4400 fax: 415.291.4401 www.ppic.org PPIC SACRAMENTO CENTER Senator Office Building 1121 L Street, Suite 801 Sacramento, California 95814 phone: 916.440.1120 fax: 916.440.1121" ["_edit_lock"]=> string(13) "1628192108:34" } ["___content":protected]=> string(102) "

R 711LHR

" ["_permalink":protected]=> string(103) "https://www.ppic.org/publication/unauthorized-immigrants-in-california-estimates-for-counties/r_711lhr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8804) ["ID"]=> int(8804) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:41:00" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(4156) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 711LHR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_711lhr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_711LHR.pdf" ["wpmf_size"]=> string(6) "951763" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(57651) "Unauthorized Immigrants in California Estimates for Counties July 2011 Laura E. Hill and Hans P. Johnson with research support from David Ezekiel and Joseph M. Hayes Supported with funding from the Silicon Valley Community Foundation http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 2 Summary California has more unauthorized immigrants than any other state, about 2.6 million of the nation’s 11 million; they make up 7 percent of the total California population and 9 percent of the state’s labor force. For decades, unauthorized immigrants have been a part of California: in many industries in the economy and in rural and urban communities. But recent and comprehensive information about t he numbers and location of this population within California —at the county and sub -county level—does not exist. That this wide information gap exists i s doubly surprising given the amount of energy spent and attention paid to this issue by policymakers and the public over those same decades. This report is the first to use a new source of administrative data at the local level to produce comprehensive and systematic sub -state estimates of the unauthorized immigrant population in California . We find that unauthorized immigrants live in every county in the state, primarily but not only in highly agricultural or highly urban areas. As in the country as a whole, unauthorized workers here reside not just in traditional immigrant communities, but have found homes throughout all regions of the state. Contents Summary 2 Tables 4 Figures 5 Introduction 6 Counting California’s Unauthorized Immigrants 7 Obstacles to Counting the Unauthorized 8 Combining New Administrative Data and Residual Method Data 9 Methodology 12 Where in California Do Unauthorized Immigrants Live? 16 Unauthorized Immigrant Population Time Trends 17 Unauthorized Immigrants Live in Zip Codes t hroughout the State 21 Conclusion 28 References 29 About the Authors 30 Acknowledgments 30 A technical appendix to this paper is available on the PPIC website: http://www.ppic.org/content/pubs/other/711LHR_appendix.pdf http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 4 Tables Table 1 Estimates of unauthorized immigrants in California and the United States (millions) 9 Table 2 Stepwise regression estimates 14 Table 3 Estimates of California county unauthorized immigrant populations (2008) 16 Table 4 Estimates of county unauthorized immigrant populations: preferred model; implied by ITIN tax filings; implied by new noncitizens 19 http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 5 Figures Figure 1 Unauthorized immigrants in California and all other states, selected years 7 Figure 2 Tax returns filed with ITINs 11 Figure 3 Correlations between ITIN tax filers and unauthorized immigrant estimates, all states 12 Figure 4a Estimates of unauthorized immigrants in California, by zip code 22 Figure 4b Estimates of unauthorized immigrants in California, percent of population, by zip code 23 Figure 5a Estimates of unauthorized immigrants, Los Angeles County zip codes 24 Figure 5b Estimates of unauthorized immigrants, percent of population, Los Angeles County zip codes 25 Figure 6a Estimates of unauthorized immigrants, San Francisco Bay Area zip codes 26 Figure 6b Estimates of unauthorized immigrants, percent of population, San Francisco Bay Area zip codes 27 http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 6 Introduction The most basic measures of any population group, its size and location, remain elusive when it comes to unauthorized immigrants in California . Although information about the size of the unauthorized immigrant population in the state—about 2.6 million in 2010 1— is easy to find , no current, comprehensive estimates of this population exist for smaller areas such as counties or cities . Nor are they available for any other state. This information gap creates a problem for local, state, and even federal authorities as they try to evaluate this population and create policies affecting them. It is especially problematic in California, where the unauthorized immigrant population is so large and its location so geographically diverse. In this work, we seek to close this knowledge gap at the sub -state level for California . P roducing sub -state estimates of unauthorized immigrants is challenging because immigrants cannot be counted directly. In this work, we take a unique set of administrative data, in this case IRS tax return data for unauthorized immigrants, and then model the data’s relationship to estimates of unauthor ized immigrants for the states using regression analysis. Because the IRS tax data is available at the zip code level, we ca n use the observed relationship between IRS tax data and state populations of unauthorized immigrants to estimate unauthorized immigrant population s for counties and sub-county areas. Our methodology is one common for demographers estimating population gr owth; both the Census Bureau and the California Department of Finance , for example, use administrative data on births and housing to estimate size and change for large population sets such as age and ethnic groups (although not unauthorized immigrants). A further advantage to our method is that IRS data are released annually. Thus, estimates can be both updated and rep licated across other states that have large unauthorized immigrant populations. We first explain our data and methodology in detail , then pr esent our results. 1 Passel and Cohn (2010 ); Hoefer, Rytina, and Baker ( 2010). http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 7 Counting California’s Unauthorized Immigrants The Pew Hispanic Center (PHC), the Department of Homeland Security (DHS), and Warren ( 2011) produce careful estimates of the size of the unauthorized immigrant population at the state and national level . Because there are no national or state level surveys that obtain the documentation status of the foreign -born residing in the United States, these three sources provide our best indirect estimates of the number and distribution of unauthorized immigrants . This population is in the midst of a major shift. After many years of increases, the number of California’s unauthorized immigrants has remained stable or even declined slightly recently . 2 At the same time, the number living in other st ates has increased substantially compared to California (Figure 1). In 1980, approximately half the nation’s unauthorized immigrants lived in the state, but that share had fallen substantially , to about 26 percent, by 2008. FIGURE 1 Unauthorized immigrants in California and all other states, selected years SOURCE S: Passel and Woodward ( 1984); Warren (2011 ). NOTE: All years except 1980 are from Warren (2011 ). There are other signs of change in national settlement patterns: states with the highest rates of growth of unauthorized immigrants in recent years are not the traditional ones : Mississippi, Alabama, and South Carolina (Warren 2011). Further, the greatest numerical gains in recent years have been in Texas, Florida, and North Carolina . More recently, estimates suggest some states are losing unauthorized immigrant populations —including California, which had 250,000 fewer in 2009 than in 2008, according to DHS 2 Passel and Cohn (2011 ); Warren (2011); Department of H omeland Security ( 2011). 1.0 1.5 2.5 3.0 3.0 2.1 2.1 6.0 8.2 8.5 0 1 2 3 4 5 6 7 8 9 1980 1990200020052008 Unauthorized Immigrants (millions) California All other states http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 8 estimates and 100,000 fewer in 2009 than in 2007, according to the PHC. (The PHC estimate d difference is not statistically significant.) Within California, settlement patterns may be changing as well but until now, there was no way of gauging these. Obstacles to Counting the Unauthorized Because the Census and national population surveys place their primary focus on generating full participation, they do not ask foreign -born participants to reveal their immigration status for fear that they will not participate . The Census, the American Community Survey (ACS) , and the Current Population Survey (CPS) do ask respondents who identify themselves as foreign -born to state their country of birth, their date of arrival, and whether or not they are naturalized U.S. citizens . For those not naturalized, however, no further survey questions provide the detail necessary to determine if they are legally resident (either permanently or temporarily) or unauthorized. A few other surveys come closer to providing counts of the unauthorized than do the census, CPS, or ACS, but none can do the job completely. T he National Agricultural Worker Survey ask s foreign -born participants to state their official immigration status, but their respondents are all agricultural workers, and therefore not representative of the full population of unauthorized immigrants . (Although more than half of all agricultural workers are estimated to be unauthorized, only 4 percent of the nation’s total unauthorized population is employed in agriculture , Passel (2009) finds). The Cal ifornia Health Interview Survey asks respondents to indicate if they are citizens or legal permanent residents, but does not differentiate between temporary visa holders and unauthorized immigrants. Th e New Immigrant Survey provides retrospective information on prior legal status of immigrants who eventually gain legal permanent residency , but the s ample size is too small to pinpoint locations and the se unauthorized immigrants may not be representative of all unauthorized immigrants in all ways . 3 Thus, estimates of unauthorized immigrants at the national and state level are produced indirectly, using what is commonly referred to as a residual technique. We rely on three sources for California data on the unauthorize d that use this technique —PHC, DHS, and Warren (2011). Each is computed using slightly different data and slightly different variants on a residual method approach. To count the foreign- born population, t he PHC estimates use Current Population Survey (CPS) data while the DHS and Warren estimates use the American Community Survey ( ACS). Next, each subtracts estimates of the legal foreign - born residents from the counts of the foreign- born in the CPS or ACS . The remaining, or residual, foreign - born comprise th e estimates of unauthorized immigrants . The three estimates vary somewhat in how they determine the legally resident population of the foreign- born, but all three combine administrative counts of legal admissions with standard demographic techniques. 4 (For more detailed descriptions, see Passel and Cohn 2010, Passel 2007, Hoefer, Rytina, and Baker 2011, and Warren 2011.) The estimates computed by these three are generally in close agreement . 3 Hill, Lofstrom, and Hayes ( 2010) demonstrated many similarities. 4 Warren does not estimate the legal population fully each year. Instead, he estimates legal foreign -born arrivals each year and uses demographic techniques to esti mate changes in the legal population from one year to the next. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 9 TABLE 1 Estimates of unauthorized immigrants in California and the United States (millions) 2000 2005 2006 2007 2008 2009 2010 California Warren 2.6 2.9 2.9 2.9 2.9 PHC 2.3 2.7 2.8 2.7 2.6 2.6 DHS 2.5 2.8 2.8 2.8 2.9 2.6 2.6 Total U.S. Warren 8.6 10.9 11.1 11.2 11.2 PHC 8.4 11.1 12.0 11.6 11.1 11.2 DHS 8.5 10.5 11.3 11.8 11.6 10.8 10.8 SOURCES: Warren (2011); Passel and Cohn ( 2010, 2011); DHS (2010 , 2011 ). In California, credible efforts to estimate some sub-state unauthorized populations have been undertaken, but none are as current as our data and none derive from annually updated, independent, administrative data. Pastor and Marcelli in 2004 published estimates of Mexican unauthorized immigrants in California Public Use Microdata Areas (PUMA s) for 1990 and 2000. 5 These were based on two survey s of Mexican immigrants in Los Angeles County, with the results used to determine characteristics of Mexican unauthorized immigrants. These characteristics were in turn used to assign probabilities of unauthorized status among Mexican immigrant s in the pop ulation data. Heer and Passel (1987) demonstrated that estimates similar in type to those of Pastor and Marcelli ( based on survey data that are applied to population data) matched well with estimates derived by interpolating from estimates based on the res idual method. The Census Bureau did estimate counts of unauthorized immigrants for California counties in the 1980s, but to our knowledge, these estimates were not published, with the exception of Los Angeles County (Heer and Passel 1987) , and were later d iscontinued. More recently, Fortuny, Capps, and Passel (2007) published estimates for five metropolitan areas in California (and 25 such areas nationwide) for the years 2000 and 2003 –2004. In addition, P aral and Associates publishes on its web page estimat es of unauthorized immigrants by U.S. congressional districts for 2000 and 2005. They apportion the Passel estimates to congressional districts based on demographic correlates. Combining New Administrative Data and Residual Method Data As described above, residual method estimates for unauthorized immigrants at the state level are in wide agreement and are generally believed to be the best source of information about unauthorized immigrants in the United States. We use these reliable estimates as the basis from which to derive new estimates for regions, counties, and smaller geographic areas within California , combining them with new administrative data to do so . Because the administrative data we use are not in wide use (despite having been collected since 1996), and have never before been used for this purpose, we devote this section to explaining them and their usefulness for our estimates. 5 PUMAs are larger than zip codes and census tracts. In sparsely populated regions, PUMAs can span many counties, but within dense counties, such as Los Angeles, there may be several. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 10 Since 1996, unauthorized immigrants have been permitted to file federal tax returns using a unique identifier , the Individual Taxpayer Identification Number, or ITIN. Immigrants and the native- born with the right to work use social security numbers (SSN) when filing tax returns, but those without work authorization do not have valid SSNs and should use ITINs instead. 6 Th e IRS has made counts of ITIN filers by zip code publicly available for tax years 2000– 2007 (which correspond to calendar years 200 1–2008). As we show, these counts of ITIN filers are a basis from which to estimate unauthorized immigrants in counties and s ub-count areas in California. Even if they have worked in the United States without proper authorization, unauthorized immigrants are nevertheless required under federal law to file tax returns . Some e stimates suggest that about half do so ( Immigration Policy Center 2011 ; Pastor et al . 2010; Hinojosa -Ojeda 2010), but others show the share is much higher: a recent PPIC report found that over 8 0 percent of unauthorized immigrants reported having filed federal income taxes in the year prior to earning legal permanent residence (Hill, Lofstrom, and Hayes 2010). The Social Security Administration ’s chief actuary estimated that about 75 percent of unauthorized immigrants have payroll taxes withheld (Porter 2005). Unautho rized immigrants have many incentives to file tax returns . First, some who have had taxes withheld by their employers throughout the year would be eligible for tax refunds and might use an ITIN to claim that money . Second, if an unauthorized immigrant ulti mately does become a legal permanent resident and receives an SSN, he or she can link any social security earnings withheld under the ITIN to earnings under the new SSN and have them counted toward later benefits. Third, even unauthorized immigrants who have not had taxes withheld (being self -employed or paid in cash, for example) might file tax returns to establish a positive paper trail for the future: should comprehensive immigration reform ever become a reality, a clear record of employment and tax payments are factors likely to increase an unauthorized immigrant’s chances of attaining legal status . T he majority of ITIN users are unauthorized immigrants, as we explain below. If a former ITIN filer is ever granted the legal ability to work, he or she shou ld begin using his or her assigned SSN immediately. Further, a nyone who is authorized to work in the United States, such as those on temporary work visas (e.g. H -1B or foreign students with work authorization s) are required to apply for and file federal ta xes with an SSN . By December 2008, the IRS had issued more than 14 million ITINs. Not all of these ITINs are used on tax returns filed domestically; s ome are filed from abroad. Many may have been retired after an immigrant was issued a valid SSN. Some may never have been used to file taxes (they may have been obtained to open a bank account, for example). Some may no longer be in use because the ITIN holder no longer files taxes, either b y choice or because he or she no longer is required to do so ( having insufficient income or no longer liv ing in United States). The number of tax returns filed with ITINs and that use U.S. addresses has increased dramatically since 1996. 7 The years for which we have data —2001 to 2008—reveal that many tax filers in California were early users of the ITIN, making up 40 percent of ITIN filers nationally in 2001 , although this fell to 30 percent of filers in by 2008 (Figure 2). Other states have seen similar but smaller changes in their share of ITIN tax filers. 6 ITINs are not a valid proof of employment eligibility; ITIN tax filers could not receive federal stimulus tax rebates, nor ar e they eligible for the Ea rned Income Tax Credit (EITC), which requires a filer, spouse, and child to have valid SSNs. 7 Anomalous increases in ITIN filings were investigated and in some cases, those records were deleted from the totals reported here. Texas and Georgia both had single -year increases in a few zip codes that were larger than the entire zip code p opulation, so we kept the growth in ITIN numbers for those f our zip codes at the level recorded in the previous year. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 11 For example, Ill inois had 4.7 percent of all filers in 2001 and 6.1 percent in 2008. Like California, ITIN tax filers in New York also declined in share of ITIN filers (from 6.5 percent to 5.2 percent ). North Carolina ’s share rose (from 1.8 percent to 3.7 percent over the same period). These changes are largely consistent with the residual estimates of unauthorized immigrants in these states. For example, California’s share of the nation’s unauthorized population fell from 43 percent in 2000 to 35 percent in 2008, according to the Warren estimates. FIGURE 2 Tax returns filed with ITINs SOURCE: Authors’ calculations of ITIN numbers . NOTE: Beginning in 2006, dependents with ITINs were also recorded. In California, nearly 6 percent of tax filers used an ITIN in 2008 (929,000) , up from 2 percent in 200 1. These include primary filers , spouses, or dependents . Although not all ITIN filers were unauthorized, the number of ITIN filers was 36 percent of the number of estimated unauthorized immigrants. A lthough ITIN tax filing data may serve as a good proxy for unauthorized immigrants, they can not provide a precise count , for two reasons . First, not all unauthorized immigrants file taxes, and among those that do, not all use the ITIN . Some may file instead using a false, fraudulent SSN, or an SSN issued many years ago that did not permit work 8 or one that no longer does . Second, not all ITIN filers are unauthorized immigrants . H owever, once we exclude tax filers from abroad, the vast majority of ITIN filers do appear to be unautho rized. When we examine ITIN tax filers with U.S. filing addresses, we find that 90 percent in 2008 include wages. Only unauthorized workers would file tax retu rns with wages and an ITIN ; authorized workers with wages would file tax returns using SS Ns. Therefore, it is safe to conclude that the vast majority of ITIN fil ers are unauthorized workers . 8 These “no work” SSNs are no longer issued. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 2001 2002200320042005200620072008 Taxes filed with ITINs (millions) Total California Florida New York Texas http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 12 Methodology Using ITIN data to estimate the distribution of unauthorized immigrants within California (or any state) depends on a strong relationship between the ITIN numbers and existing estimates of unauthorized immigrants. ITIN numbers will not include all unauthorized immigrants, nor will they include only unauthorized immigrants . But to use ITINs as a basis from which to scale local area estimates to match state total estimates we need the correlation between ITIN counts and unauthorized estimates to b e high. And indeed, we find ITINs are highly correlated with independently derived state estimat es of unauthorized immigrants. We calculate d correlations for each year of available ITIN data with the Warren and PHC estimates (Figure 3 ). Each point in the figure represents that year’s correlations for the 50 states and the District of Columbia. FIGURE 3 Correlations between ITIN tax filers and unauthorized immigrant estimates, all states SOURCES: Authors’ calculations from ITIN counts ; Warren (2011); Passel and Cohn ( 2011, 2010). NOTE: 2008 ITIN correlations for Passel and Cohn use 2009 data; DHS estimates are available only for the 10 states with the largest unauthorized populations . The Warren estimates have a correlation value of nearly 0.98 with the ITIN tax filing records (see Appendix Table A 1 for the full set of Warren estimates) . The PHC estimates use ranges for states with small populations, which may partially explain their lower correlation , which is still very high: in excess of 0.955 for all years . In the PHC estimates, those of 32 states are derived using multiple years of data because of the very small CPS data sample of likely unauthorized immigrants in those states—fewer than 50 in each year ( Passel and Cohn, 2011) . Altogether, these extremely high correlations give us great confidence that ITIN filings are an excellent indicator of the unauthorized immigrant population. We also examined the relationship between the estimates of unauthorized immigrants and ITIN data filed with wages attached (W -2s) , ITIN taxing filings prepared by a paid tax preparer, and ITINs not filed as a n on resident. Each of these has the potential to be more highly correlated with unauthorized immigrants than the overall ITIN filing rates. ITI N returns filed with wages are almost definitely unauthorized immigrant tax filers (all other filers with the legal right to work should have SSNs), b ut not all ITIN unauthorized tax filers 0.9450.95 0.955 0.96 0.965 0.97 0.975 0.98 0.985 0.99 2001 2002200320042005200620072008 Correlations Warren Pew Hispanic Center http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 13 will have wages attached (e.g. those who file but are paid cash or are self -employed ). The vast majority of ITIN filers use paid preparers , and do so at much higher rate than tax filers in general . This may be because unauthorized immigrant tax filers may prefer to pay for help in filing correctly if their motivation is to document earnings and tax paying ; or it may simply be that unauthorized tax filers who use paid preparers are m ore likely to learn about ITINs than if they do not . 9 Unauthorized immigrants should also be more likely to file as residents (rather tha n using the nonresident 1040NR form) because nonresidents can not claim child tax credits and because filing as a nonresident might do less to establish one’s intent to ultimately legalize. 10 However, we found n one of these other measures of ITIN tax fil ers were as consistently highly correlated as the correlations with the overall ITIN tax filings for the states. Given the high correlation between ITIN filers and estimates of the unauthorized for the 50 states and D .C., we could use the simple ratio of unauthorized immigrants to ITIN filers as the factor to estimate local populations of unauthorized immigrants . However, we know this ratio varies across states and time , and so suspect that it might also vary within the state. We use regression analyses to account for this cross-state and intra- state variation . Differences in the rate of ITIN usage by a state’s unauthorized population may be related to variation s in the characteristics of employment and earnings among unauthorized immigrants, their demograp hic characteristics , or infrastructure available to support immigrant tax filing, among others . Using the Warren estimates, ITIN data, the ACS (for the 2006 and 2008 models) , and the 2000 c ensus (for the 2001 model ), we use weighted least squares (WLS) to estimate the following regression model of the ITIN coverage rate for each state (s) and each year ( t), with proportionate weight applied to the estimated size of the undocumented population . We restrict the population to foreign -born residents : ��������� ������������ ����������������� ���� = ������α + ������β+ ������γ+������ In the equation, X represents a matrix of demographic characteristics of the immigrant population in each state and each year . It includes age, proportion Latino, and proportion born in Central America; w e do not include the proportion born in Mexico because that is so closely correlated with the proportion s that are Latino and Central America n-born . Employment characteristics for the immigrant population are represented by the mat rix W and include the share employed in construction, the share employed in restaurants, proportion self - employed, and proportion not in the labor force. Tax filing characteristics are represented by the matrix Z and include filing as married, being a new tax filer, and filing using a paid preparer. Because we have only 51 observations ( 50 states and D.C. ), we restricted our possible covariates to just a few (Table 2). We want these relationships to be able to vary, first because the increase in ITIN usage was so great during th e interval we studie d (Figure 2), and second, because we do not fully understand why some states seemed to have a higher percentage of unauthorized immigrants filing income taxes with ITINs . We therefore allowed the variables that en tered into this equation to change across time . We used a backward elimination stepwise regression method, removing the least significant variable from the model one at a time until all variables met a predetermined threshold of significance. In this case our threshold wa s a p-value of 0 .10. Regressions were run separately for each year. 9 Filers claiming the EITC also use paid preparers at a very high rate. In California, 76 percent of EITC filers used paid preparers in 2006 (Danielson 2010). 10 1040NR tax forms can be filed from U.S. addresses by those who are only temporarily residing in the U.S. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 14 In developing our model, we considered different population bases to use to estimate the ratio of ITIN tax filers to Warren estimates , and settled on the foreign -born as the base population of each state . 11 We discarded some of the variables that were determined to be significant based on the state- level models because the range of values for those variables for counties in California was far outside the range of values for states. For example, the share of the foreign -born population working in agriculture ranges from l percent to 25 percent in the 50 state s and D .C., but in California counties, the high end is much greater: Tulare County has 61 percent of its foreign -bo rn labor force working in agriculture, and many other California counties were above 40 percent. We considered models in which we predicted the ITIN numbers, with the Warren estimates on the right hand side of the equation. We also predicted the logged IT IN numbers. Both of these models resulted in state -level computed estimates that had a poorer fit than the ratio models—that is, our predictions for states were not as close to the actual Warren estimates as in our final model . Further , we estimated our models using the PHC estimates as robustness check. Because correlations are lower (Figure 3) and the estimates are not available for all year s, we prefer the Warren estimates . (The models estimated with the PHC data are available on request.) Our final mo del varied for each year. We report those for 2001, 2006, and 2008 below; Table 2 reports all of the variables that we allowed to enter into the stepwise regression . As noted above, only those variables which were estimated with a p value of 0.10 were ultimately included in the regression . We have the most confidence in the county estimates for 2008 . TABLE 2 Stepwise regression e stimates 2001 2006 2008 R squared 0.701 0.593 0.618 hettest 0.012 0.501 0.692 Coeff p-value Coeff p-value Coeff p-value Age 0 –17 Age 35–54 0.215721 0.038 – Age 55+ 0.186604 0.006 Proportion Latino Born in Central America -0.1601 0.000 -0.41705 0.000 -0.5378 0.000 Construction Restaurants Self-employed Not in the labor force Filed taxes as married -0.11164 0.000 -0.39599 0.000 -0.48947 0.000 New tax filer Filed using paid preparer 0.065578 0.064 Constant 0.015552 0.718 0.489259 0.000 0.598122 0.000 SOURCES: Authors’ calculations using ITIN, ACS, 2000 census, and Warren and Associates data . 11 We also considered models using re cently arrived foreign -born noncitizens as the base population . Those models had so much more variation within California counties than across the 50 states and strained the ability of regression to make useful estimates. Expanding the sample more widely to include the entire population resulted in a poorer fitting model as well. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 15 In each of the years 2001, 2006, and 2008, the share of the foreign -born population from Central America and the share of ITIN tax filers that filed tax returns as married was very important in scaling the ITIN returns to m atch the state -level estimates of unauthorized immigrants . Also important for 2001 was the share of the foreign -born population between ages 35 and 54, and 55 or o lder, as was the share of ITIN tax filers who used a paid tax preparer. Applying these coeff icient estimates for the states to ITIN tax -filer and ACS data for the counties, we computed a county - (or regional -) level count of unauthorized immigrants. 12 These were then totaled and scaled to match the estimate of unauthorized immigrants for the state in that year . Our final step was to scale these local estimates back down to the zip code level, using the distribution of ITIN filings filed by zip code within that county . (We could not use our model and ACS data in the same way because ACS data are not available for that small geography.) 12 We totaled zip codes to the county or regional level. Many zip codes span two (or more) counties, and are allocated to counti es based on 2000 census block populations ( Kneebone 2008). http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 16 Where in California Do Unauthoriz ed Immigrants Live? For 2008, we find unauthorized immigrants residing in all counties (or county groups) throughout the state. 13 Unauthorized immigrant s are found in major urban areas, agricultural regions, and places in between. According to our estimates, their raw numbers range from just over 1,000 in the Del Norte/Siskiyou/Modoc/ Lasse n county grouping to just under 1 million in Los Angeles County (Table 3). In general, unauthorized immigrants make up small but notable shares of county populations. In only four counties or county groupings do they make up more than 10 percent of the total population. In 22 counties, mostly rural and mountainous, but also including Sacramento, unauthorized immigrants make up less than 5 percent of the population. Not surprisingly, the most populous counties have the largest populations of unauthorized immigrants . TABLE 3 E stimates of California county unauthorized i mmigrant populations (2008) County Total population (2008 ACS estimate) Unauthorized immigrants Population estimate % of county total population Alameda 1,475,000 124,000 8.4% Amador, Calaveras, Tuolomne, Mariposa, Alpine, Mono, Inyo 191,000 2,500 1.4% Butte 220,000 4,000 1.8% Colusa, Glenn, Tehema, Trinity 124,000 10,000 8.3% Contra Costa 1,029,000 79,000 7.7% Del Norte, Siskiyou, Modoc, Lassen 118,000 1,000 1.0% El Dorado 176,000 4,000 2.2% Fresno 909,000 49,000 5.3% Humboldt 129,000 2,000 1.6% Imperial 164,000 21,000 12.8% Kern 801,000 46,000 5.7% Kings 150,000 9,000 5.8% Los Angeles 9,860,000 916,000 9.3% Madera 149,000 12,000 7.7% Marin 249,000 14,000 5.6% Mendocino, Lake 151,000 8,000 5.0% Merced 246,000 22,000 9.1% Monterey, San Benito 463,000 62,000 13.5% Napa 134,000 16,000 12.0% 13 Because the ACS is a sample, not all counties have large enough populations to be reported separately in it. Because of these ACS sample size restrictions, we report unauthorized immigrant estimates for 34 counties and 7 county groups rather than 58 counties for both 2001 and 2008. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 17 County Total population (2008 ACS estimate) Unauthorized immigrants Population estimate % of county total population Orange 3,010,000 289,000 9.6% Placer 342,000 8,000 2.3% Plumas, Sierra, Nevada 120,000 2,000 1.5% Riverside 2,101,000 146,000 7.0% Sacramento 1,394,000 65,000 4.6% San Bernardino 2,015,000 150,000 7.5% San Diego 3,002,000 198,000 6.6% San Francisco 809,000 30,000 3.7% San Joaquin 673,000 54,000 8.0% San Luis Obispo 266,000 9,000 3.5% San Mateo 712,000 55,000 7.8% Santa Barbara 405,000 37,000 9.0% Santa Clara 1,764,000 180,000 10.2% Santa Cruz 253,000 21,000 8.2% Shasta 180,000 1,000 0.6% Solano 407,000 24,000 6.0% Sonoma 467,000 41,000 8.8% Stanislaus 511,000 39,000 7.6% Sutter, Yuba 165,000 9,000 5.6% Tulare 426,000 29,000 6.8% Ventura 798,000 74,000 9.3% Yolo 198,000 12,000 6.2% Total 36,756,000 2,876,000 7.8% SOURCE S: Authors’ calculations ; ACS . Unauthorized Immigrant Population Time Trends In this section, we present our estimates for 2001 and 2008 together. We expect ed that our method and model s would be more reliable for years when ITIN numbers were more commonly used than the years when ITINs were new and less likely to be used by unauthorized immigrants. However, the fit for our model that predicted the ratio of ITIN filings to the Warren unauthorized estimates was actually slightly better in 2001 than 2008 (Table 2), despite the fact that, as Figure 2 illustrated , there was a dramatic uptick in use of ITINs among tax filers. We also found that from 2001 to 2008 the number of zip codes with ITIN tax filers increased. Taken individually, the estimates for the single years seem reasonable. Although our 2001 model fits well, we are still cautious about our results from years before 2008 mainly because a smaller share of unauthorized immigrants was filing ITIN returns in the earlier years; when w e examine the change from 2001 to 2008, we have less certainty about the prior years . We find that for many of the small county and small county groupings, the growth in unauthorized immigrants that is implied from our estimates is perhaps TABLE 3 (continued) http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 18 implausible. Therefore, the 2001 estimates we present are only for those counties with 2008 populations of 200,000 or greater and those in which our 200 1 estimate of unauthorized immigrants was greater than 10,000. We offer two benchmarks , for comparison only . The first is the simple distribution of the estimated number of unauthorized immigrants using the ITIN counts for counties from the administrative data. Comparing our model estimates to the ITIN results gives a sense of how our model may be an improvement over simply scaling the administrative tax data . The second benchmark is the distribution of the estimate of unauthorized immig rants using the distribution of the state’s new noncitizens (arrived within the last 20 years ) to counties. This is one way to allocate the reputable state estimates to sub -state areas (but not a method employed by any of those who compute those residual methods) . Our method, because its underlying data are available every year, for all zip codes nationwide, and because it does not rely on any other allocation or estimation (with the exception of state -level estimates) , is the best methodology available gi ven the current data constraints . http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 19 TABLE 4 E stimate s of county u nauthorized immigrant populations : preferred model ; implied by ITIN tax filings; implied by new non citizen s* 2001 2008 Change 2001 to 2008 County County population (2008 ACS) Model ITIN filers New non-citizens Model ITIN filers New non-citizens Model ITIN filers New non-citizens Alameda 1,475,000 163,000 144,000 121,000 124,000 116,000 116,000 -39,000 -28,000 -5,000 Amador, Caleveras, Tuolmne, Mariposa, Alpine, Mono, Inyo 191,000 2,500 3,000 2,000 Butte 220,000 4,000 4,000 6,000 Colusa, Glenn, Tehema, Trinity 124,000 10,000 10,000 6,000 Contra Costa 1,029,000 63,000 63,000 50,000 79,000 71,000 71,000 16,000 8,000 21,000 Del Norte, Siskiyou, Modoc, Lassen 118,000 1,000 1,000 1,500 El Dorado 176,000 4,000 4,000 4,000 Fresno 909,000 30,000 25,000 63,000 49,000 51,000 72,000 19,000 26,000 9,000 Humboldt 129,000 2,000 2,500 1,000 Imperial 164,000 21,000 18,000 14,000 Kern 801,000 21,000 18,000 36,000 46,000 46,000 53,000 25,000 28,000 17,000 Kings 150,000 9,000 9,000 12,000 Los Angeles 9,860,000 924,000 948,000 1,069,000 916,000 894,000 987,000 -8,000 -54,000 -82,000 Madera 149,000 12,000 12,000 12,000 Marin 249,000 16,000 19,000 12,000 14,000 14,000 12,000 -2,000 -5,000 0 Mendocino, Lake 151,000 8,000 8,000 5,000 Merced 246,000 15,000 12,000 16,000 22,000 24,000 24,000 7,000 12,000 8,000 Monterey, San Benito 463,000 39,000 37,000 43,000 62,000 73,000 48,000 23,000 36,000 5,000 Napa 134,000 16,000 15,000 12,000 Orange 3,010,000 349,000 387,000 273,000 289,000 323,000 267,000 -60,000 -64,000 -6,000 Placer 342,000 8,000 7,000 10,000 Plumas, Sierra, Nevada 120,000 2,000 2,000 500 http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 20 2001 2008 Change 2001 to 2008 County County population (2008 ACS) Model ITIN filers New non-citizens Model ITIN filers New non-citizens Model ITIN filers New non-citizens Riverside 2,101,000 78,000 75,000 85,000 146,000 145,000 145,000 68,000 70,000 60,000 Sacramento 1,394,000 42,000 36,000 60,000 65,000 64,000 80,000 23,000 28,000 20,000 San Bernardino 2,015,000 100,000 99,000 95,000 150,000 141,000 125,000 50,000 42,000 30,000 San Diego 3,002,000 189,000 165,000 175,000 198,000 184,000 177,000 9,000 19,000 2,000 San Francisco 809,000 42,000 51,000 64,000 30,000 33,000 65,000 -12,000 -18,000 1,000 San Joaquin 673,000 31,000 27,000 35,000 54,000 51,000 43,000 23,000 24,000 8,000 San Luis Obispo 266,000 9,000 10,000 9,000 San Mateo 712,000 64,000 71,000 60,000 55,000 55,000 65,000 -9,000 -16,000 5,000 Santa Barbara 405,000 37,000 36,000 29,000 37,000 39,000 35,000 0 3,000 6,000 Santa Clara 1,764,000 241,000 246,000 182,000 180,000 185,000 190,000 -61,000 -61,000 8,000 Santa Cruz 253,000 15,000 17,000 16,000 21,000 24,000 17,000 6,000 7,000 1,000 Shasta 180,000 1,000 1,000 2,000 Solano 407,000 16,000 15,000 16,000 24,000 23,000 22,000 8,000 8,000 6,000 Sonoma 467,000 42,000 43,000 22,000 41,000 43,000 26,000 -1,000 0 4,000 Stanislaus 511,000 23,000 22,000 24,000 39,000 38,000 28,000 16,000 16,000 4,000 Sutter, Yuba 165,000 9,000 9,000 10,000 Tulare 426,000 33,000 33,000 29,000 29,000 32,000 33,000 -4,000 -1,000 4,000 Ventura 798,000 48,000 49,000 45,000 74,000 83,000 54,000 26,000 34,000 9,000 Yolo 198,000 12,000 12,000 12,000 Total 36,757,000 2,711,000 2,711,000 2,711,000 2,876,000 2,876,000 2,876,000 165,000 165,000 165,000 SOURCE S: Authors’ calculations ; 2008 ACS . * New noncitizens are non- naturalized foreign-born who arrived in the previous 20 years (ACS 2008). TABLE 4 (continued) http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 21 According to our estimates, the population of unauthorized immigrants in Los Angeles County was 924,000 in 2001 and declined very slightly to 916,000 by 2008. Two other sources have also estimated this population for similar years; the Los Angeles Family and Neighborhood Survey of 2000 f ound that there we re about 664,000 unauthorized adults in Los Angeles County. Using the residual method, Fortuny, Capps, and Passel (2007) estimate d about 937,000 in the same year, 2000 . 14 Our estimates are much closer to those that use the residual approach . Similarly, c hanges over time in large counties appear to be smaller proportionally, according to our model , than unauthorized immigrant population changes in the smaller counties . Unauthorized Immigrants L ive in Zip Codes t hroughout the State Within counties, we are able to match, approximate ly, the unauthorized immigrant population to the zip codes of residence. As we described above, however, we cannot use our model to directly estimate unauthorized immigrant residents in zip codes . Instead, we take our county -level estimates (derived from our models ), and allocate the unauthorized immigrants based on the distribution of ITIN tax return filers by zip code within the county . There are three potential problems with this approach. First, counts of fewer than 10 ITIN tax filers at the zip code level were suppressed by the IRS , so county totals of unauthorized immigrants cannot be allocated to that zip code. However, the number of zip codes with 10 or more ITIN tax filers has risen rapidly in our data years. In 2001, 54 percent of zip codes with any tax filers had ITIN tax filers; by 2008, th at share had risen over two thirds, 67 percent. Second, some zip codes are actually points, such as post office boxes, or office buildings . These are not mapped, but the data from them are included in our county estimates (Tables 3 and 4 ). Third, because tax filers may use a work address or an address other than a residence, we find that in a few zip codes we predict higher numbers of unauthorized residents than there are total residents . Th ese are few: in 2008, we found nine , defined as zip codes where the total population was fewer than 1, 000 and the per centage of unauthorized residents was greater than 35 percent. In our maps that display the percentage of zip code residents that are unaut horized immigrants, we do not show levels over 15 percent , and so do not expect that these nine zip codes dramatically alter the visual presentation of our results. O ur methods clearly cannot predict unauthorized immigrants residing in the state’s zip codes with exact precision . For that reason, we present our zip code results in ranges, rather in specific number or tabular form . In addition, we do not separate zip codes with zero unauthorized immigrants from zip codes with just a few unauthorized immigrants because of the IRS data suppression issue. Maps of the state by zip code reveal unauthorized immigrants residing in some very highly concentrated pockets throughout the state, but also located in some places of relative isolation . Throughout the state, we find zip codes with more than 5 ,000 unauthorized immigrant residents well outside highly urbanized areas (Figure 4 a). When we consider the unauthorized as a percentage of the populat ion, we find many zip codes where 15 percent of the population is unauthorized acros s even more diffuse and diverse geographies (Figure 4b). Maps for Los Angeles County (Figures 5a and 5b) and for the San Francisco Bay area (Figures 6a and 6b) are provided to illustrate the patterns that emerge from estimating sub -country distributions . All maps reflect 2008 data. 14 Fortuny et al. (2007) also provide estimates for Los Angeles County in 2003-04 (1,000,000), Orange County PMSA (245,000 in 2000), and Riverside/San Bernardino PMSA (175,000 in 2000). http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 22 FIGURE 4A Estimates of unauthorized immigrants in California , by zip c ode SOURCE: Authors’ calculations using ITIN and ACS data. NOTE : Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 23 FIGURE 4B Estimates of u nauthorized immigrants in California, percent of population, by zip code SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 24 FIGURE 5A Estimates of unauthorized immigrants, Los Angeles County zip codes SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 25 FIGURE 5B Estimates of u nauthorized immigrants , percent of population, Los Angeles County zip codes SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 26 FIGURE 6A Estimates of unauthorized immigrants , San Francisco Bay Area zip codes SOURCE: Authors’ calculations using ITIN and ACS dat a. NOTE: Areas in white indicate no population . http://www.ppic.org/main/home.asp Unauthorized Immigrants in California 27 FIGURE 6B Estimates of u nauthorized immigrants , percent of population, San Francisco Bay Area zip codes SOURCE: Authors’ calculations using ITIN and ACS data. NOTE: Areas in white indicate no population. http://www.ppic.org/main/home.asp 28 Conclusion We have developed comprehensive sub-state estimates of unauthorized immigrants for California —perhaps the most important state to understand when considering i mmigration issue s. Our estimates are based on administrative data— income tax returns by unauthorized immigrant s—available for local areas. Prior to the ir availability, the best es timates about where in the state unauthorized im migrants reside were limited to larger levels of geography and are now either outdated or avail able only for subsets of th is population. As with any estimates of unauthorized immigrants, these numbers are subject to uncertainty . However, t he administrative data we rely on is highly correlated with independently developed residual estimates of state unauthorized immigrant populations . We take further comfort in our results for Los Angeles County, which are consistent with other estimates derived from the residual method. We expect that as the percentage of unauthorized immigrants using ITIN numbers on their tax returns increases across the country , our method could be used to compute similar sub -state estimates in other locations. Currently, it is reasonable to attempt to do so with states with large populations of unauthorized immigrants who appear to be using ITINs in large number s, such as Texas, New York, Illinois, and Arizona.     http://www.ppic.org/main/home.asp 29 References Danielson,  Caroline.  2010.  “The  Earned  Income  Tax  Credit  in  California.”  Just  the  Facts.  Public  Policy  Institute  of  California.   Available  at www.ppic.org/main/publication.asp?i=925 .  Fortuny,  Karina,  Randy  Capps,  and  Jeffrey  S.  Passel.  2007.  “The  Characteristics  of  Unauthorized  Immigrants  in   California,  Los  Angeles  County,  and  the  United  States.”  Washington  DC:  Urban  Institute.   Goldman,  Dana P.,  James  P.  Sm ith ,  and  Neeraj  Sood.  2006.  “Immigrants  and the  Cost  of  Medical  Care.”  Health  Affairs  25  (6).   Heer,  David  M.,  and  Jeffrey  S.  Passel.  1987.  “Comparison  of Two  Methods  for  Estimating  the Number  of  Undocumented   Mexican  Adults  in  Los  Angeles  County.”  International  Migration  Review  21  (4).    Hill,  Laura  E.,  Magnus  Lofstrom , and  Jo seph M.  Hayes.  2010.  Immigrant  Legalization:  Assessing  the  Labor  Market  Effects.  San  Francisco:  Public Policy Institute  of  California.  Available  at www.ppic.org/main/publication.asp?i=869.   Hinojosa ‐Ojeda, Raul.  2010.  “Raising  the  Floor  for  American  Workers:  The Economic  Benefits  of  Comprehensive   Immigration  Reform.”  Washington  DC:  Center  for  American  Progress  and Immigration  Policy  Center.   Hoefer,  Michael,  Nancy  Rytina,  and  Bryan  C.  Bak er.  2011.  “Estimates  of  the  Unauthorized  Immigrant  Population   Residing  in  the  United  States:  January  2010.”  Population  Estimates,  Office of  Immigration  Statistics,  Department  of   Homeland  Security.    Immigration  Policy  Center.  2011.  “Unauthorized  Immigrants  Pay  Taxes,  Too:  Estimates  of  the  State  and  Local  Taxes  Paid   by  Unauthorized  Immi grant Households.”   Internal  Revenue  Service. SPEC  Tax  Data.  Available  through  Brookings  EITC  interactive  ( www.brookings.edu ).  Kneebone,  Elizabeth.  2008.  “Bridging  the  Gap:  Refundable  Tax  Credits  in  Metropolitan  and  Rural  America.”  Earned   Income  Tax  Credit  Series,  Metropolitan  Policy  Program  at  Brookings.  Washington,  DC:  Brookings  Institute.   Passel,  Jeffrey.  2007.  “Unauthorized  Immigrants  in  the  Unit ed  Stat es: Estimates,  Methods,  and  Characteristics.”  OECD   Social,  Employment  and  Migration  Working  Paper.    Passel,  Jeffrey  S.,  and  Karen  A.  Woodrow.  1984.  “Geographic  Distribution  of  Undocumented  Immigrants:  Estimates  of   Undocumented  Aliens  Counted  in  the  1980  Census  by  State.”  International  Migration  Review  18  (3).    Passel,  Jeffrey  S.,  and  D’Vera  Cohn.  2011.  “Unauthorized  Immigrant  Popu lation:  Na tional and  State  Trends,  2010.”   Washington  DC:  Pew  Hispanic  Center.  Passel,  Jeffrey  S.,  and  D’Vera  Cohn.  2010.  “U.S.  Unauthorized  Immigration  Flows  Are  Down  Sharply  since  Mid ‐Decade.”   Washington  DC:  Pew  Hispanic  Center.  Passel,  Jeffrey  S.,  and  D’Vera  Cohn.  2009.  “A  Portrait  of  Unauthorized  Immigrants  in  the  United  States.”  Washington,   DC:  Pew  Hispanic  Center.  Pastor,  Manuel ,   Justin  Scoggins,  Jennifer  Tran,  and  Rhonda  Ortiz.  2010.  “The  Economic  Benefits  of  Immigrant   Authorization  in  California.”  Los  Angeles:  Center  for  the  Study  of Immigrant  Integration,  University  of  Southern   California.   Pastor  Jr.,  Manuel,  and  Enrico  A.  Marcelli.  2004.  “Somewhere  Over the  Rainbow?  African  Americans,  Unauthorized  Mexican  Immigrati o n,  and  Coalition  Building.”  In  The  Impact  of  Immigration  on  African  Americans,  ed.  Steven   Schulman  (Piscataway,  NJ:  Transaction  Publishers).   Porter,  Eduardo.  2005. “Illegal  Immigrants  Are  Bolstering  Social  Security  with  Billions.”  New  York  Times,  April  5.    Rob  Paral  and  Associates.  “Undocumented  Immigrants  in  Congressional  Districts.”  Available  at   www.robparal.com/MapPage.html?map=14&type=G . Accessed  May 5,  2011  (Google  Earth  plug‐in  required).   Warren,  Robert.  2011.  “Annual  Estimates  of  the  Unauthorized  I mmigrant  Population  in  the  United  States,    by  State:  1990  to 2008.”  Working  paper,  Public  Policy Institute  of  California.  Available  at  http://www.ppic.org/main /publication.asp?i=992.  http://www.ppic.org/main/home.asp Title 30 About the Author s Laura E. Hill is a policy fellow at the Public Policy Institute of California. Her research interests include immigrants, immigration, race and ethnicity, and youth. She has been a research associate at The SPHERE Institute and a National Institute of Aging postdoctoral fellow. Sh e holds a Ph.D. in demography from the University of California, Berkeley. Hans Johnson is a senior policy fellow at the Public Policy Institute of California. His research focuses on the dynamics of population change in California and policy implications of the state’s changing demography. At PPIC, he has conducted research on international and domestic migration, population projections, housing, and higher education. Before joining PPIC as a research fellow, he was senior demographer at the California Re search Bureau, where he conducted research on population issues for the state legislature and the governor’s office. He has also worked as a demographer at the California Department of Finance, specializing in population projections. He holds a Ph.D. in de mography from the University of California, Berkeley. Acknowledgments The authors wish to thank Robert Warren and Elizabeth Kneebone and for their help with the data. Technical reviews of the method and paper by Jeffrey Passel and Manuel Pastor are much appreciated. We also thank Helen Lee, Kim Belshé, Robert Gleeson, and Abby Cook for multiple reviews and conversations, and Richard Greene for his skillful wordsmithing. PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors John E. Bryson, Chair Retired Chairman and CEO Edison International Mark Baldassare President and CEO Public Policy Institute of California Ruben Barrales President and CEO San Diego Regional Chamber of Commerce María Blanco Vice President, Civic Engagement California Community Foundation Gary K. Hart Former State Senator and Secretary of Education State of California Robert M. Hertzberg Partner Mayer Brown LLP Walter B. Hewlett Director Center for Computer Assisted Research in the Humanities Donna Lucas Chief Executive Officer Lucas Public Affairs David Mas Masumoto Author and farmer Steven A. Merksamer Senior Partner Nielsen, Merksamer, Parrinello, Gross & Leoni, LLP Constance L. Rice Co- Director The Advancement Project Thomas C. Sutton Retired Chairman and CEO Pacific Life Insurance Company http:// The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research on major economic, social, and political issues. The institute’s goal is to rais e public awareness and to give elected representatives and other decisionmakers a more informed basis for developing policies and programs. The institute’s research focuses on the underlying forces shaping California’s future, cutting across a wide range of public poli cy concerns, including economic development, education, environment and resources, governance, population, public finance, and social and health policy. PPIC is a private operating foundation. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. PPIC was established in 1994 with an endowment from William R. Hewlett. Mark Baldassare is President and Chief Executive Officer of PPIC. John E. Bryson is Chair of the Board of Directors. Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notic e is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. © 2011 Public Policy Institute of California All rights reserved. San Francisco, CA PUBLIC POLICY INSTITUTE OF CALIFORNIA 500 Washington Street, Suite 600 San Francisco, California 94111 phone: 415.291.4400 fax: 415.291.4401 www.ppic.org PPIC SACRAMENTO CENTER Senator Office Building 1121 L Street, Suite 801 Sacramento, California 95814 phone: 916.440.1120 fax: 916.440.1121" ["_edit_lock"]=> string(13) "1628192108:34" ["post_date_gmt"]=> string(19) "2017-05-20 09:41:00" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_711lhr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:41:00" ["post_modified_gmt"]=> string(19) "2017-05-20 09:41:00" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_711LHR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }