というNBER論文が上がっている（ungated版へのリンクがある著者の一人のページ）。原題は「Should Friday be the New Saturday? Hours Worked and Hours Wanted」で、著者はGregor Jarosch、Laura Pilossoph、Anthony Swaminathan（いずれもデューク大）。
以下はその要旨。
This paper investigates self-reported wedges between how much people work and how much they want to work, at their current wage. More than two-thirds of full-time workers in German survey data are overworked—actual hours exceed desired hours. We combine this evidence with a simple model of labor supply to assess the welfare consequences of tighter weekly hours limits via willingness-to-pay calculations. According to counterfactuals, the optimal length of the workweek in Germany is 37 hours. Introducing such a cap would raise welfare by .8-1.6% of GDP. The gains from a shortened workweek are largest for workers who are married, female, white collar, middle aged, and high income. An extended analysis integrates a non-constant wage-hours relationship, falling capital returns, and a shrinking tax base.
（拙訳）
本稿は、人々がどれだけ働いているか、と、彼らが現在の賃金でどれだけ働きたいか、についての自己申告の差を調べた。ドイツの調査データにおけるフルタイム労働者の2/3以上は働き過ぎ――実際の労働時間が望む労働時間を超過――であった。我々はこの実証結果を労働供給の簡単なモデルと組み合わせ、週労働時間に対するより厳しい制限の厚生面での帰結を、喜んで支払う意思の計算を通じて評価した*1。反実仮想によれば、ドイツにおける最適な週労働時間は37時間であった。そうした上限を導入すると、厚生はGDPの0.8-1.6%上昇することになる。週労働時間の短縮による便益は、高所得のホワイトカラーの中年既婚女性で最も大きかった。さらなる分析では、賃金と時間の非定常な関係、低下する資本の収益率、および税基盤の縮小を統合する*2。
*1:本文では「The model allows us to convert the observed and counterfactual wedges into dollars via simple willingness to pay (WTP) calculations.」と記述している。
*2:本文では以下のように説明している：
「A final section extensively addresses two considerations related to the resource feasibility of the counterfactual allocations we consider. First, we ask whether firms would actually be willing to hire workers at the counterfactual hours we consider at an unchanged wage rate. For an answer, we follow Bick et al. (2022) and investigate the empirical relationship between hours and wages in the cross-section in our data. We then apply the resulting schedule to calculate the surplus or shortfall we should expect relative to the constant wage–hours benchmark. Similar to Bick et al. (2022), we find a declining relationship between hours and wages across much of the relevant part of the support. This implies that our calculations based on a constant wage-hours relationship understate the gains from a shortened workweek. The gains are larger yet if workers sort positively on hours in the cross-section.
Second, we offer a way of integrating into the analysis losses from reducing working hours that are arguably “external” to the survey respondents. Specifically, we allow for both the income earned by capital owners and the tax base to contract when aggregate hours fall. To do this, we work with a simple CES production function and calculate the total income reduction associated with the change in total effective hours implied by a shortened workweek.
Under these two extensions, maximum welfare is obtained at 36 hours, with net gains of 3.88% of GDP. These large gains, however, primarily reflect that the cross-sectional exercise `a la Bick et al. (2022) suggests significant productivity gains from reducing hours. A more conservative alternative shuts this down and consequently predicts net welfare losses, albeit small, at any hours cap below 43. The takeaway is that a shorter workweek at constant wages delivers sizable welfare gains to workers in Germany. Whether these are offset, from an economy-wide perspective, by a smaller tax base and lower returns to capital depends on whether labor productivity rises as hours fall.」
（Bick et al. (2022)はHours and Wages* | The Quarterly Journal of Economics | Oxford Academic）
