Asset Growth and the Cross-Section of Stock Returns

Source: Cooper, M. J., Gulen, H. & Schill, M. J. (2008). Journal of Finance 63(4), 1609–1651.

TL;DR

Firms that grow their total assets quickly earn low subsequent returns, and slow-growers earn

high returns. The asset-growth effect is large, robust across size groups, and one of the most

economically significant cross-sectional predictors — it underlies the modern "investment" factor (CMA

in the Fama-French five-factor model).

What anomaly it documents

A negative relation between the rate of total-asset growth and future returns: aggressive expanders

(via capex, acquisitions, issuance) underperform conservative firms. The signal uses the whole balance

sheet, not a single line item.

How it is constructed

Sorting variable: year-over-year growth in total assets.

Universe: US common stocks; the effect survives in large caps, not just microcaps.

Portfolio: long low-asset-growth, short high-asset-growth; value- or equal-weighted, rebalanced

annually.

Evidence and replication

Portfolio	Result	Source
Low − high asset growth	Large positive spread, robust to FF/momentum controls	this paper
By size	Present even among large stocks	this paper

Why it might work

Overinvestment / empire-building that the market initially rewards then corrects (mispricing).

Investment-based asset pricing: firms invest more when their cost of capital (expected return)

is low — a risk-based reading consistent with q-theory.

Limitations and risks

Distinguishing mispricing from a rational investment factor is hard.

Annual rebalancing on accounting data needs point-in-time discipline; costs apply.

Key references

Cooper, M., Gulen, H. & Schill, M. (2008) — Asset Growth and the Cross-Section of Stock Returns — Journal of Finance

Titman, S., Wei, K. C. & Xie, F. (2004) — Capital Investments and Stock Returns — Journal of Financial and Quantitative Analysis

Fama, E. & French, K. (2015) — A Five-Factor Asset Pricing Model — Journal of Financial Economics

Reference replication on ConvexPi

An open, verified replication of this strategy is maintained at convexpi/replications. It recomputes the strategy from underlying building blocks and scores it out of sample (the McLean & Pontiff test):

Period	Annualized Sharpe
In-sample (pre-2008)	+0.71
Out-of-sample (≥ 2008)	+0.03
Last 10 years	-0.05

Verdict: dormant. Run it on live data in Colab · view the code