Dani Sandler

One was career navy, the other a structural engineer. Great-grandfathers were a janitor, a lawyer, a steel worker, and one who worked in a department store.

CSWEP Newsletter Issue I is out now. Take a moment to read it at chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.aeaweb.org/content/file?id=24297

Post a pic you took, no context, to bring some zen to the feed.

This paper examines how multi-unit firms’ life-cycle stages affect analyst forecast accuracy. While prior studies focus on the firm-level life cycle, we utilize the Census data and focus on the establishment level. We find that analyst forecast accuracy is lower for multi-unit firms whose establishments are in different life-cycle stages than those in the same life-cycle stage. This finding suggests that the forecasting difficulty of more diversified firms can be attributed to the different life-cycle stages of each establishment. We also find that for firms whose units are in different stages, analyst forecast accuracy is lower if the establishments in earlier stages are larger (i.e., generate more revenue) than those in later stages. As a comparison, we estimate the life-cycle stages using firms’ segment classifications in their 10-K filings. We find that analysts’ forecast accuracy is lower when firms report fewer segments than the number of establishments, suggesting that aggregating more establishments for segment reporting could complicate analysts’ forecasting. To our knowledge, this is the first study that focuses on the establishmentlevel life cycle. This study highlights that firm-level life cycles should not be taken without caution, as aggregating multiple units’ life cycles may be misleading. In order to provide better forecasts to investors, analysts should have a deeper understanding of firms’ subunits, especially when the establishments are in different life-cycle stages.

These tables present the correlation between establishment proportions and segment proportions in the lifecycle stages measured using the Census Bureau’s establishment data and firms’ 10-K segment data, respectively. Panel A shows the Pearson correlation between the proportions of firms’ segments in five lifecycle stages using the 10-K segment sales (SEG_STGn_SALES) and the proportions of firms’ establishments in the five stages using the establishment TVS (STGn_TVS) where n= 1, 2, 3, 4, and 5. Panel B shows the correlation between the proportions using the 10-K segment capex (SEG_STGn_CAPEX) and the proportions using the establishment CAPEX (STGn_CAPEX). Panel C shows the correlation between the proportions using the 10-K segment employment (SEG_STGn_EMP) and the proportions using the establishment EMP (STGn_EMP). ***, **, * indicate statistical significance at the 1%, 5%, and 10% levels (two-tailed), respectively.

Using Census data, this paper shows that analyst forecasts are less accurate when multi-unit firms’ establishments span different life-cycle stages.

Establishment-level life cycle and analysts’ forecasts Using Census data, we show that analyst forecasts are less accurate when multi-unit firms’ establishments span different life-cycle stages.

New Census Working Paper: "Establishment-level life cycle and analysts’ forecasts" by Sudipta Basu, Xin Dai, Caroline Lee

www.census.gov/library/work...

Looking for a way to get more involved with the Society of Government Economists? Nominations for the board of directors are being accepted through March 3rd. It’s a great group! www.sge-econ.org/about/electi...

My daughter makes hard boiled eggs for her lunches. Usually she just marks them with colors, but she left them on Sunday for me to put away.

He is old and just wants to cuddle and be high up.

Another week, another cool data product release from LEHD and Census. This week, we released new PSEO data for schools from North Carolina and BYU-Idaho, bringing our total count of institutions to 952.

Check out the data here: lehd.ces.census.gov/applications...

Using administrative data on 17 million U.S. business applications linked to outcomes, we compare potential entrants’ expectations about employer entry and first-year employment with realizations. On average, applicants overestimate employment, mainly because many expect to enter but do not. Among those who expect and achieve entry, employment is typically underestimated. Expected employment predicts entry and realized employment, but conditional on entry realized employment rises less than one-for-one with expectations. Expectation errors are highly heterogeneous and systematically related to application characteristics and local economic conditions, and they predict near-term employment outcomes. A parsimonious model with heterogeneous priors, learning, and pre-entry selection rationalizes these patterns.

Notes: Dashed curves are 12-month centered moving averages (12-MA). Red vertical line indicates March 2020; grey vertical line indicates March 2019.

Linking 17M business applications to outcomes, we examine how potential entrants’ expectations about entry and initial size line up with reality.

Expectations versus Reality in Business Formation Linking 17M business applications to outcomes, we examine how potential entrants’ expectations about entry and initial size line up with reality.

New Census Working Paper: "Expectations versus Reality in Business Formation" by Emin Dinlersoz and Yueyuan Ma
www.census.gov/library/work...

Over the last decade, research on labor market adjustment following local demand shocks has expanded to explore a wide variety of measured shocks. However, the worker adjustments observed in response to these shocks are not always consistent across studies. We create a harmonized set of annual commuting-zone-level shocks following the major approaches in the literature to investigate these differences. As one might expect, shocks of different types exhibit different geographic and temporal patterns and are generally weakly correlated with each other. We find they also generate different employment and migration responses, with trade-related shocks showing little response on either margin, while more general Bartik-style shocks are associated with economically meaningful changes in both employment and migration.

Notes: The figures show results from an event study regression for the employment growth shock. Panel A shows the results for every year in our data, while Panel B shows the result when we separate this analysis into Census years. Underlying employment data are the CZ-level data described in Section 2.

Different local labor demand shocks lead to different employment and migration responses. Bartik shocks are associated with significant effects on both margins.

A Shock by Any Other Name? Reconsidering the Impacts of Local Demand Shocks Different local labor demand shocks lead to different employment and migration responses. Bartik shocks are associated with significant effects on both margins.

New Census Working Paper: "A Shock by Any Other Name? Reconsidering the Impacts of Local Demand Shocks" by Sean Bassler, Kevin Rinz, David Wasser, and Abigail Wozniak

www.census.gov/library/work...

Women’s college attendance delivers wage and job-quality gains that grow over the life cycle, alongside improvements in children’s early-life health, from Na'ama Shenhav and Danielle H. Sandler www.nber.org/papers/w34767

We did, but did not find a difference in fertility around the cutoff

Notes: This fgure presents the estimated efect (with 90% confdence intervals) of being born just before the school entry cutof on mid-career job amenities, rescaled as the willingness-to-pay for the amenity value of one more year of education. The efect of the discontinuity on each job amenity is estimated using occupation-level averages calculated from the American Working Conditions Survey (Maestas et al., 2017), described in Appendix Section B2. Each estimate (and standard error) is then multiplied by the willingness-to-pay for the amenity from Maestas et al. (2023), and divided by the efect of the cutof on years of education (shown in Table A.7) to obtain the wage-equivalent return to education from each amenity (and standard error). The estimate labeled ”all” represents the sum across all twelve amenities. Estimates are weighted using a triangular kernel and use a bandwidth of 70 days around the cutof. Standard errors are clustered by day of birth relative to the school entry cutof. Source: California birth records 2007–2017.

We also see increased levels of valuable job amenities from increased college attendance, which means the value of increased college attendance is higher than wages alone would indicate.

Notes: This figure presents the estimated effect (with 95% confidence intervals) of being born just before the school entry cutoff on average earnings (Panel A), the likelihood of working (Panel B), and household earnings (Panel C) at each age between 20–34. Earnings panel includes women born between 1977–1987, who satisfy other sample requirements. Estimates are weighted using a triangular kernel with a bandwidth of 70 days around the cutoff. Standard errors are clustered on day of birth relative to the school entry cutoff. Source: Linked IRS–Census administrative earnings records. See text for details.

The effects of increased college attendance growth over the lifecycle, so that we see bigger effects for women in mid-career, both on earnings and employment and on the birthweight of children born to mid-career mothers.

Life-Cycle Effects of Women's Education on their Careers and Children Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

This was also released as an NBER working paper this morning: www.nber.org/papers/w34767

We use the sharp RD in education from entry cutoffs has been shown in previous research to increase college attendance to look at longer run outcomes, including earnings, employment, and infant health.

We study the causal effect of women's education on their wages, non-wage job amenities, and spillovers to children. Using a regression discontinuity at the school entry birthdate cutoff, we find that women born just before the cutoff are more likely to complete some college, and experience multi-dimensional career gains that grow over the life cycle: greater employment and earnings, as well as more professional and higher-status jobs, more socially meaningful work, and better working conditions. Children’s early-life health and prenatal inputs improve in tandem with career improvements, consistent with professional advances spurring—not hindering— infant investments. Career gains are concentrated in jobs that require exactly some college, the same schooling margin shifted by the cutoff, which indicates that increased post-secondary education is the primary channel for these effects. Together, the results show that women's college attendance generates large career returns—from both wages and amenities—that strengthen over time and produce meaningful benefits for children.

Notes: This figure presents average earnings (panel A) and employment rates (panel B) for women born around the school entry cutoff. Outcomes are residualized with (recentered) year of birth fixed efects. Average outcomes (shown on the y-axis) are reported for 10 equally-sized bins of birth dates on each side of the cutofh. The sample includes women ages 16–40, born between 1977–1987, who satisfy other sample requirements. Source: Linked IRS–Census administrative earnings records. See text for details.

Increased women’s college attendance delivers wage and job-quality gains that grow over the life cycle, alongside improvements in children’s early-life health.

Life-Cycle Effects of Women's Education on their Careers and Children Increased women’s college attendance delivers wage and job-quality gains that grow over the life cycle, alongside improvements in children’s early-life health.

New Census Working Paper: "Life-Cycle Effects of Women's Education on their Careers and Children" by Na'ama Shenhav and Danielle H. Sandler
www.census.gov/library/work...

Immigrant students who attend U.S. colleges are disproportionately employed in either large firms—especially multinationals—or small firms and self-employment. Using linked Census and longitudinal employment data, we trace the jobs taken by college students in 2000 during the 2001-20 period and evaluate four mechanisms shaping sector and firm size placement: geographic clustering, degree specialization, firm capabilities/visas, and ethnic self-employment specialization. Degree fields predict large firm and MNE placement, while ethnic specialization explains small firm sorting. Immigrant students who remain in the U.S. earn more than their native peers, suggesting the segmentation reflects productive sorting rather than blocked opportunity.

Notes: Column 1 provides the baseline multi-factor analysis. This regression and Columns 3-6 are weighted by a triple interaction of log employment in the ethnic group, the sector, and the firm size category. Column 2 weights observations using log count of immigrant students from place of birth who are working during 2006-10 in the sector and firm size cell. Columns 3 and 4 quantify factors explaining job placements among large and small firms, respectively. Columns 5 and 6 quantify factors explaining job placements within and outside of multi-nationals, respectively. Explanatory regressors in Columns 3-6 are kept the same as baseline analysis. Regressions report robust standard errors. Disclosure conducted under FSRDC Project Number 2766. CBDRB-FY24P2766-R11782. FY25-P2766-R12183.

Immigrant students who attend US colleges take jobs in the largest and smallest firms and show strong future earnings.

Positioned at Extremes: Future Job Placements of Immigrant Students at U.S. Colleges Immigrant students who attend US colleges take jobs in the largest and smallest firms and show strong future earnings.

New Census Working Paper: "Positioned at Extremes: Future Job Placements of Immigrant Students at U.S. Colleges" by Francis Dillon, Sar Pekkala Kerr, William R. Kerr, and Andrew Wang
www.census.gov/library/work...

We characterize the careers of minimum wage workers by merging SIPP panels covering 19922016 into the LEHD. A long-run analysis shows strong earnings growth for these workers in subsequent decades, becoming indistinguishable from peers earning modestly more initially. Most of this growth is due to the steep earnings trajectories of young workers. Older workers earning minimum wages show a modest dip in earnings at that moment compared to earlier and later periods. Increases in state minimum wages do not significantly alter the future careers of workers who are on the minimum wage when the increases occur.

Notes: Figure reports the 1990-2021 LEHD earnings and state-indexed earnings percentiles of individuals who earned minimum wages during the 1994-1998 period as measured by the SIPP. Values for LEHD earnings are in 2021 dollars and winsorized at 5% and 95% levels per year. The left-side panels describe career profiles for those earning $0-5.50, $5.51-$7.00, and $7.01+ per hour in 1994-1998 (nominal values). The right-side panels separate the future career trajectories of workers earning $0-5.50 in 1994-1998 by age (15-29, 30-49) and family income levels (separated at $3000 per month). The 22 states present in the LEHD sample: AZ, CA, CO, CT, DE, KS, MD, ME, MO, MT, ND, NM, NV, OH, PA, SC, SD, TN, VA, WA, WI, and WY. Disclosure conducted under FSRDC Project Number 2766. CBDRB-FY24-P2766-R11520. CBDRB-FY25-P2766-R11999.

Workers earning minimum wages typically show stronger future careers, and state minimum wage policies do not significantly alter these relative trajectories.

Careers of Minimum Wage Workers Workers earning minimum wages typically show stronger future careers, and state minimum wage policies do not significantly alter these relative trajectories.

New Census Working Paper: "Careers of Minimum Wage Workers" by Sari Pekkala Kerr, William R. Kerr, and Louis Maiden
www.census.gov/library/work...

The U.S. Census Bureau’s Person Identification Validation System facilitates anonymous linkages between survey and administrative records by assigning Protected Identification Keys (PIKs) to person records. While PIK assignment is generally accurate, some person records are not successfully assigned a PIK, which can lead to sample selection bias in analyses of linked data. Using the American Community Survey (ACS) and the Current Population Survey Annual Social and Economic Supplement (CPS ASEC) between 2005 and 2022, we corroborate and extend existing findings on the drivers of PIK assignment, showing that the rate of PIK assignment varies widely across socio-demographic subgroups. Using earnings as a test case, we then show that limiting a survey sample of wage earners to person records with PIKs or successful linkages to W2 wage records tends to overestimate self-reported wage earnings, on average, indicative of linkage-induced selection bias. In a validation exercise, we demonstrate that reweighting methods, such as inverse probability weighting or entropy balancing, can mitigate this bias.

Notes: These figures show differences in average wage earnings between the target sample of non-zero wage earners aged 15-64 and employed in the government or private sector and a restricted sample of those assigned a PIK (missing PIK bias) or linked to a W-2 record (linkage bias). Each panel compares baseline to adjusted bias estimates, where adjustments are produced using a limited or full set of covariates. All estimates are produced using survey-specific person weights. 2020 is omitted due to high survey non-response. Shaded regions represent 95% confidence intervals. Lines represent linear interpolation of survey year statistics. Data sources: ACS, CPS ASEC, IRS W-2s.

Reweighting methods like entropy balancing mitigate sample selection bias in analyses of household surveys that condition on linkable person records.

Non-Random Assignment of Individual Identifiers and Selection into Linked Data: Implications for Research Reweighting methods like entropy balancing mitigate sample selection bias in analyses of household surveys that condition on linkable person records.

New Census Working Paper: "Non-Random Assignment of Individual Identifiers and Selection into Linked Data: Implications for Research" by Kyle Raze, Nicole Perales, and Liana Christin Landivar
www.census.gov/library/work...

Week 1 - Black History Month 2026
1. You may have seen this image of Black men voting during Reconstruction, and assumed it is a depiction of voting after the passage of the 15th Amendment (which outlaws denying the right to vote based on race).

This housing program helped kids escape poverty — by changing who they befriended : Planet Money In the 1990s, Congress created HOPE VI, a program that demolished old public housing projects and replaced them with more up-to-date ones. But the program went further than just improving public housi...

Planet Money coverage: www.npr.org/2026/01/28/n...

We study whether low-economic-mobility neighborhoods can be transformed into high-mobility areas by analyzing the HOPE VI program, which invested $17 billion to revitalize 262 distressed public housing developments. We estimate the program’s impacts using a matched differencein-differences design, comparing outcomes in revitalized developments to observably similar control developments using anonymized tax records. HOPE VI reduced neighborhood poverty rates by attracting higher-income families to revitalized neighborhoods, but had no causal impact on the earnings of adults living in public housing units. Children raised in revitalized public housing units earn more, are more likely to attend college, and are less likely to be incarcerated. Using a movers exposure design and sibling comparisons, we show that these improvements were driven by changes in neighborhoods’ causal effects on children’s outcomes. The improvements in neighborhood causal effects were driven in large part by changes in social interaction: HOPE VI increased interaction between public housing residents and peers in surrounding neighborhoods and increased earnings more for subgroups with higher-income peers. Many low-income families in the U.S. currently live in neighborhoods that are as socially isolated as the HOPE VI developments were prior to revitalization. We conclude that it is feasible to create high-opportunity neighborhoods and that connecting socially isolated areas to surrounding communities is a cost-effective approach to doing so.

Notes: This figure plots estimates of the difference in adults’ mean incomes between HOPE VI and matched control public housing projects j years after a HOPE VI grant. The figure plots estimates of β j from the stacked difference-in-differences specification in equation (10). All specifcations control for project fxed effects and grant-year-by-calendar-year fxed effects. Each series presents estimates from a separate regression with a different outcome variable. For the green series, we construct a repeated crosssectional dataset consisting of all adults living in the physical footprint of the original public housing project in each year, and the outcome variable is the average household income among this population. For the orange series, we construct an individual-level panel following the original adult residents of the public housing projects—adults who lived in the projects one year prior to the grant year—and the outcome variable is the average household income among this population. For the navy series, we construct a repeated cross-sectional dataset consisting of adults living in public housing in each year. For each individual, we measure the difference between their current household income and their household income one year prior to moving into public housing, and the outcome is the average value of this difference. All specifcations are estimated using weighted least squares with inverse propensity scores as weights. Standard errors are clustered by project, and the vertical bars denote 95% confdence intervals.

HOPE VI revitalization cut neighborhood poverty and boosted children’s adult earnings and college via reduced social isolation—without raising adults’ earnings.

Creating High-Opportunity Neighborhoods: Evidence from the HOPE VI Program HOPE VI revitalization cut neighborhood poverty and boosted children’s adult earnings and college via reduced social isolation—without raising adults’ earnings.

New Census Working Paper: "Creating High-Opportunity Neighborhoods: Evidence from the HOPE VI Program" by Raj Chetty, Rebecca Diamond, Thomas B. Foster, Lawrence Katz, Sonya R. Porter, Matthew Staiger, and Laura Tach

www.census.gov/library/work...