Tazwar Bin Saleh Building for impact.

Does it matter how a developing country's agriculture grows, by using more inputs or by using them better? This paper combines the USDA's decomposition of agricultural growth into total factor productivity and four inputs with FAO nitrogen balances and UN national accounts in a gap-free panel of 102 countries for 1971–2023, 75 of them low- or middle-income, and tests hypotheses written down before estimation. In the typical developing country, inputs supplied three-quarters of output growth. Over five-year periods, growth in fertiliser and feed is accompanied by a rising nitrogen surplus, whereas productivity growth is accompanied by rising nitrogen use efficiency, a larger fall in emissions per unit of output and an unchanged surplus; these associations are contemporaneous, partly a matter of accounting and partly reversed in the next period. A higher surplus is associated with faster later productivity growth only below about 15 kg N/ha. Across developing countries the surplus rises with income and turns down only among high-income countries, yet eleven of 29 developing countries with a substantial surplus reduced it after 2000 while raising output, regardless of income and without slower growth. Two hypotheses are not supported: a decade of productivity-led growth is not followed by faster non-farm growth than input-led growth, and greater exposure to world urea prices does not predict larger output losses. In Bangladesh the surplus rose from 138 to 169 kg N/ha while Vietnam's fell; whether Bangladesh's productivity growth has turned negative depends on an undocumented choice of cost shares in the published accounts.

This paper estimates what a marginal dollar of cash was worth to the shareholders of 5,173 United States non-financial firms over 2006–2022, and whose cash was worth more, using Faulkender–Wang return regressions on 39,445 firm-years that include firms later delisted. A dollar is worth 1.29 dollars at the sample means. With firm characteristics measured before the year's return, cash is worth more at firms that made a loss in the previous year, lack an investment-grade rating, are recently listed or have high Tobin's q, and less at the smallest firms and at firms whose debt is mostly short-term; the Kaplan–Zingales, Whited–Wu and size–age indices carry no reliable information. End-of-year sorting produces two further results that vanish under start-of-year sorting. An apparent premium on cash held through funding stress comes from the single recovery year 2009; stress has no reliable effect. The loss and size results depend on the return benchmark.