<![CDATA[Objective: The unemployment rate reported in official statistics does not fully capture labor market conditions as it overlooks hidden unemployment. These individuals, who appear to be employed, are not optimally engaged in economic activities. This study aims to nowcast hidden unemployment in Indonesia using high-frequency big data from Google Trends, thereby addressing the limitations of official statistics and providing more adaptive labor market indicators within the broader context of employment dynamics and policy evaluation. The research is empirical in nature. Design/Methods/Approach: The study employs time-series data, combining official statistics from the National Labor Force Survey (SAKERNAS) with search query data from Google Trends. Three econometric models — MIDAS, U-MIDAS, and BMF VAR — are applied to assess their performance in nowcasting hidden unemployment. The analysis is divided into pre-pandemic and combined periods to evaluate the model’s sensitivity to structural shocks, such as the COVID-19 pandemic. Findings: The results indicate that the MIDAS model outperforms the alternatives, with the lowest forecast errors (∆RMSE = 0.3538; ∆MAPE = 0.9028%) and the highest stability in capturing hidden dynamics of unemployment. Using the best-performing model, predictions for the first semester of 2025 indicate that hidden unemployment will reach 33.14 percent, reflecting persistent vulnerabilities in the labor market structure. Originality/Value: The study contributes to labor market research by integrating high-frequency big data with econometric nowcasting methods to estimate hidden unemployment, a phenomenon often overlooked in official statistics. This approach introduces a novel application of real-time indicators to enhance the timeliness and relevance of employment monitoring in emerging economies. Practical/Policy implication: The findings underscore the importance of adaptive employment policies that address hidden unemployment as a structural issue. By providing early indicators, this study offers policymakers timely insights to design responsive interventions, reduce labor market inefficiencies, and mitigate the risks of increasing employment disparities.]]>
