,
8 tweets,
4 min read
Read on Twitter
UNEVEN GROWTH: NEW PAPER [Thread] Over the past 4 decades, gains from growth have been very unevenly distributed. Why is that? In a new paper, Ben Moll @pascualrpo & I consider the role of automation in driving the rise in inequality, inc at the v top: princeton.edu/%7Emoll/UG.pdf 1/7 
The key feature of our theory is that technology permanently affects rates of return. By raising return to capital automation results in rapid individual wealth accumulation and thus higher wealth inequality. This is related to @PikettyLeMonde 'r-g’ argument. 2/7
More concentrated holdings of wealth translate into more concentrated capital income, which drives up income inequality. We show how this is distinct from - and more powerful than - the usual compositional argument that emphasises K-income is more concentrated than L-income. 3/7
On top that, automation drives the rise in labor income inequality, as those on low and middle wages are particularly badly hit. Our model accounts for nearly two-thirds of the observed changes in relative wages with wage declines predicted for those below 80th percentile. 4/7
Furthermore avg wages (output, productivity) do not increase by much following automation because aggregate capital deepening is actually limited (recall R is permanently higher!). Aggregate gains from technical change are smaller than a standard rep agent model wld predict 5/7
We find supporting evidence for the key mechanism that links capital share, returns and inequality. More generally, our model fully characterises distributions of income and wealth in general equilibrium (in closed form!), going beyond just the aggregates and averages. 6/7
Overall the model generates highly uneven growth, as noted by @gabriel_zucman and others. The simple framework can be applied in many other settings, incl as part of larger quantitative models, making it a useful addition to macroeconomist’s toolbox (we hope!) 7/7
