The Neoclassical Model of Savings, Precautionary Behavior, and Empirical Puzzles in Development Economics

This lecture delves into the neoclassical model of savings, beginning with the inter-temporal optimization problem where individuals maximize lifetime utility of consumption, discounted over time. It introduces the Bellman equation and its derivation to yield the Euler equation, a central tenet of economics stating that the marginal utility of consumption today must equal the expected discounted marginal utility of consumption tomorrow. The discussion highlights special cases, such as quadratic utility leading to consumption as a Martingale, and the concept of prudence (positive third derivative of the utility function), which implies precautionary savings: an increase in consumption variability should lead to increased savings to smooth consumption.

The lecture then transitions to empirical tests of these theoretical predictions, notably referencing Christina Paxson's work on Thai rice farmers. Paxson's research distinguishes between permanent and transitory income shocks, positing that permanent income changes should be fully consumed, while transitory changes should be entirely saved. Using rainfall variation as an instrument for transitory income, her findings generally support the idea that people save a significant portion of transitory income. The speaker also addresses the significant challenges in accurately measuring savings, especially in developing contexts where informal savings mechanisms (e.g., grains, livestock like bullocks, informal loans) are prevalent and difficult to quantify.

Further complexities are introduced by considering credit and savings constraints. The lecture explains how these constraints modify the Euler equation, leading to a consumption function characterized by buffer stock savings: individuals consume everything if their cash on hand is below a certain threshold, and save a fraction of the excess above that threshold. Simulations illustrate how different levels of risk aversion influence consumption paths and asset accumulation, showing that risk-averse households, especially the poor without access to credit, should theoretically save a substantial amount to mitigate consumption dips during bad income years.

However, the lecture concludes by presenting a significant empirical puzzle: despite the theoretical benefits and the availability of free savings accounts in field experiments (e.g., in Kenya, Uganda, Malawi, Chile), very few people actually adopt or actively use these accounts. This observed behavior contradicts the neoclassical model's predictions, suggesting a gap between rational economic theory and actual human behavior, thereby hinting at the need for more behavioral models to understand savings decisions, particularly among the poor in developing countries.

"the new classical model fails to fit the empirical facts in a pretty obvious way"

"the marginal utility of consumption today must be the marginal consumption of a majority of consumption tomorrow up to a discount Factor"

"consumption is a Martingale"

"the poor particularly dislike changes and changes in their income"

"an increase in the variability of consumption will increase the height and size of the Euler equation that means that consumption must decline and savings must increase"

"a permanent change income should be entirely consumed and transitory change in income should be entirely saved"

"it's pretty hard for households to rely on the credit Market on the well-functioning credit Market to finance consumption when they need it and therefore they should be saving a ton"

"the puzzle is that they don't"

"people say thank you very much and they don't use them at all"

"why is it so expensive like seven dollars to open a to open a savings account for people who make two dollars a day that's a lot of money proportionally"