Corporate short-termism is, at its root, an agency problem: shareholders would have managers look ahead, yet the work is costly, difficult to contract upon, and rewarded only later, so managers may undertake too little of it. How far ahead they in fact look has rarely been observed. We construct a direct firm-level measure of it from the horizon of forward-looking statements in earnings calls: the median number of months ahead to which a firm’s managers refer, which we term its attention to the future. The measure predicts subsequent investment, R&D, and innovation, and it contracts as a firm approaches financial distress. We then study how corporate governance shapes this attention, developing a model of managerial attention that nests price discipline through informed trading [Edmans and Manso, 2011] and the control–initiative tradeoff under blockholder intervention [Burkart et al., 1997]. Both predictions hold in the data: within firms, attention rises with the level and breadth of institutional ownership and falls as ownership concentrates in a single block. Consistent with the model’s intervention mechanism, the block’s negative association weakens as managerial equity rises, and liquidity-based exit alternatives do not account for it. Ownership thus shapes not only what firms ultimately do, but how far ahead their managers look before acting.

This study uses fiscal gap accounting (FGA) and generational accounting (GA) to compare US and Italian fiscal solvency. FGA and GA incorporate all government outlays and receipts, whether put on or kept off the books. FGA measures, in the form of reduced net outlays, the constant share of each future year’s GDP needed to balance the government’s intertemporal budget. GA calculates the lifetime net tax rate – lifetime taxes divided by lifetime labor earnings -- facing future generations if current generations pay nothing more, on net, than current policy mandates. Deficit accounting suggests that Italy’s 135 percent debt-to-GDP ratio places it in worse fiscal shape than the US with its 123 percent ratio. But on a fiscal-gap basis, Italy appears in far better shape regardless of the discount rate used. Based on the theoretically appropriate rate – the average real return to national wealth, the U.S. fiscal gap is 7.4%. Italy’s is 4.0%. These requisite solvency adjustments are far larger if delayed or if the UN’s more pessimistic demographic projections prevail. Neither country can expect future generations, on their own, to cover their government’s red ink. Doing so requires levying lifetime net tax rates, in each country, that exceed 100%.

Consider a linear model y = Xβ + u with u = (u₁,...,u_T) and u_t a serially correlated linear process given by u_t = ∑^∞_j=-h c_j e_t-j for a sequence of innovations e_t. Given a set of instruments Z, the “optimal GMM” estimator based on the moment condition E(Zu) = 0 is by far the most commonly used method to estimate such models. It can, however, be inconsistent unless the instruments are exogenous with respect to past innovations e_t-j for j > 0, when c_j ≠ 0 for j > 0. We propose a GLS-IV estimator valid in the general case with instruments exogenous or not, as long as they are pre-determined. It is shown to be much more efficient than GMM whether the moment condition E(Zu) = 0 is satisfied or not. We discuss issues of consistency by casting the estimators in a GMM framework with different moment conditions and instruments. To analyze the relative merits of the estimators when all are consistent, we cast them as some GLS estimator using different instruments and first-stage regression. It then becomes clear that GLS-IV involves “stronger instruments”, while GMM is more likely to be affected by issues of weak instruments. Other motivating elements and extensive simulations are presented to argue that our proposed GLS-IV estimator has better properties. As an empirical application, we revisit the extensive study of Mavroeidis et al. (2014) about the empirical relevance of the forward looking New Keynesian Phillips curve. Using the GLS-IV procedure on the same dataset, our estimates are all in the right quadrant, consistent with theoretical expectations.