Dataset Information

Nowcasting unemployment insurance claims in the time of COVID-19.

ABSTRACT: Near-term forecasts, also called nowcasts, are most challenging but also most important when the economy experiences an abrupt change. In this paper, we explore the performance of models with different information sets and data structures in order to best nowcast US initial unemployment claims in spring of 2020 in the midst of the COVID-19 pandemic. We show that the best model, particularly near the structural break in claims, is a state-level panel model that includes dummy variables to capture the variation in timing of state-of-emergency declarations. Autoregressive models perform poorly at first but catch up relatively quickly. The state-level panel model, exploiting the variation in timing of state-of-emergency declarations, also performs better than models including Google Trends. Our results suggest that in times of structural change there is a bias-variance tradeoff. Early on, simple approaches to exploit relevant information in the cross sectional dimension improve forecasts, but in later periods the efficiency of autoregressive models dominates.

SUBMITTER: Larson WD

PROVIDER: S-EPMC8846950 | biostudies-literature | 2022 Apr-Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Nowcasting unemployment insurance claims in the time of COVID-19.

Larson William D WD Sinclair Tara M TM

International journal of forecasting 20210111 2

Near-term forecasts, also called nowcasts, are most challenging but also most important when the economy experiences an abrupt change. In this paper, we explore the performance of models with different information sets and data structures in order to best nowcast US initial unemployment claims in spring of 2020 in the midst of the COVID-19 pandemic. We show that the best model, particularly near the structural break in claims, is a state-level panel model that includes dummy variables to capture ...[more]

PMID: 35185232

Similar Datasets

Project description:Forty million U.S. residents lost their jobs in the first two months of the coronavirus disease 2019 (COVID-19) pandemic. In response, the Federal Government expanded unemployment insurance benefits in both size ($600/week supplement) and scope (to include caregivers and self-employed workers). We assessed the relationship between unemployment insurance and food insecurity among people who lost their jobs during the COVID-19 pandemic in the period when the federal unemployment insurance supplement was in place. We analyzed data from the Understanding Coronavirus in America (UAC) cohort, a longitudinal survey collected by the University of Southern California Center for Economic and Social Research (CESR) every two weeks between April 1 and July 8, 2020. We limited the sample to individuals living in households earning less than $75,000 in February 2020 who lost their jobs during COVID-19. Using difference-in-differences and event study regression models, we evaluated the association between receipt of unemployment insurance and self-reported food insecurity and eating less due to financial constraints. We found that 40.5% of those living in households earning less than $75,000 and employed in February 2020 experienced unemployment during the COVID-19 pandemic. Of those who lost their jobs, 31% reported food insecurity and 33% reported eating less due to financial constraints. Food insecurity peaked in April 2020 and declined over time, but began to increase again among people receiving unemployment insurance during the final wave of the survey ahead of the federal supplement to unemployment insurance ending. Food insecurity and eating less were more common among people who were non-White, lived in lower-income households, younger, and who were sexual or gender minorities. Receipt of unemployment insurance was associated with a 4.4 percentage point (95% CI: -7.8 to -0.9 percentage points) decline in food insecurity (a 30.3% relative decline compared to the average level of food insecurity during the study period). Receipt of unemployment insurance was also associated with a 6.1 percentage point (95% CI: -9.6 to -2.7 percentage point) decline in eating less due to financial constraints (a 42% relative decline). Estimates from event study specifications revealed that reductions in food insecurity and eating less were greatest in the four-week period immediately following receipt of unemployment insurance, with no evidence of differential pre-existing trends in either outcome. We conclude that receiving unemployment insurance benefits during the period when the $600/week federal supplement was in place was associated with large reductions in food insecurity.

Project description:ImportanceThe COVID-19 pandemic has been associated with increased unemployment rates and long periods when individuals were without health insurance. Little is known about how Medicaid expansion facilitates Medicaid enrollment as a buffer to coverage loss owing to unemployment.ObjectiveTo compare changes in health insurance coverage status associated with pandemic-related unemployment among previously employed adults in states that have vs have not expanded Medicaid eligibility.Design setting and participantsThis cohort study included US adults aged 27 to 64 years who were employed at baseline in the 2020 to 2021 Current Population Survey's Annual Social and Economic Supplement, which included calendar years 2019 to 2020 (32 462 person-years). Data analyses were conducted between November 2021 and April 2022.ExposuresJob loss (ie, new unemployment) experienced during 2020.Main outcomes and measuresPrimary outcomes were coverage status (ie, uninsured status) and source of coverage (ie, employer sponsored, marketplace, and Medicaid). Using 2-way person-by-year fixed-effects regression models, changes in coverage status associated with unemployment in states that expanded Medicaid were compared with states that did not expand Medicaid. Additional analyses were performed based on prepandemic coverage status.ResultsThe cohort included 16 231 adults (mean age, 46.8 [95% CI, 46.6-47.0] years; 51.6% women). New unemployment was associated with an increase of 2.9 (95% CI, 1.1-4.6) percentage points (P = .002) in the proportion of uninsured adults in Medicaid expansion states and an increase of 10.7 (95% CI, 2.4-18.9) percentage points (P = .01) in nonexpansion states. Workers were 5.4 (95% CI, 1.9-8.9) percentage points (P = .003) more likely to enroll in Medicaid after a job loss if they lived in a Medicaid expansion state compared with workers experiencing job loss in nonexpansion states.Conclusions and relevanceIn this cohort study of US adults, unemployment-related Medicaid enrollment was more frequent in Medicaid expansion states during the COVID-19 pandemic. Medicaid expansion led to a smaller increase in uninsured adults because those who lost private insurance coverage (eg, employer sponsored) appeared more able to transition to Medicaid after job loss.

Dataset Information

Nowcasting unemployment insurance claims in the time of COVID-19.

Publications

Nowcasting unemployment insurance claims in the time of COVID-19.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets