Anirudh Krishna notes that while there is a lot of research characterising the stock of poor people, there isn’t much work on the flows into and out of poverty.

Here’s a table from a recent presentation (pdf) by Professor Krishna.

Studies that focus on changes in the level of poverty mask a lot of “steady-state” turnover. Such churning has a lot of implications for human welfare and policy effectiveness:

Because escaping poverty and falling into poverty are caused by different factors, looking only at the figure for net change will not help to develop appropriate policy responses. The net change in poverty over any period of time is obtained by subtracting one causal stream (escape) from another and separate causal stream (descent). It does not as such have an independent causal significance.