Market Timing and Capital Structure
Source: Baker, M. & Wurgler, J. (2002) · Journal of Finance 57(1), 1–32 · DOI: 10.1111/1540-6261.00414
TL;DR
A firm's **capital structure is the cumulative outcome of its past attempts to time the equity
market** — issuing stock when valuations are high and relying on debt (or repurchasing) when they are
low. Leverage is strongly negatively related to a firm's historical market valuations, and these
timing effects are very persistent: the impact of past market-to-book has a half-life of well over
ten years, contradicting any rapid rebalancing to a target ratio.
The idea
Standard theory says financing decisions either trade off costs/benefits toward a target leverage ratio
or follow a pecking order; either way, transitory swings in market value should leave no permanent
imprint. Baker & Wurgler argue the opposite. Managers exploit windows when equity is mispriced (a
practice CFOs admit to in surveys), and—crucially—firms do not subsequently rebalance these
decisions away. The result is path dependence: **low-leverage firms are those that raised funds when
their market valuations were high, and high-leverage firms are those that raised funds when valuations
were low.** Capital structure thus reflects the accumulated history of timing, not a stable target.
To capture this, they build the "external finance weighted-average" market-to-book ratio,
(M/B)_efwa. For a given firm-year it is a weighted average of the firm's past market-to-book ratios,
where each historical year's weight is that year's net external finance raised (equity plus debt issues)
relative to total external finance raised over the firm's life (eq. 3 in the paper). It therefore loads
on the valuations that prevailed when the firm was actually raising capital—the moments when leverage
could in practice be changed—rather than weighting all years equally.
Evidence
COMPUSTAT firms appearing 1968–1999); IPO dates from Jay Ritter (1968–1995) and SDC (1970–1998).
Firms are studied in "IPO time" (years since IPO), with a separate broader All-COMPUSTAT sample.
& Zingales controls), leverage is strongly negatively related to (M/B)_efwa. The relation holds
whether leverage is measured in book or market values and across control specifications.
weighted-average term still explains leverage—so temporary valuation swings produce permanent capital
structure changes; (ii) controlling for initial leverage, later valuation swings still move leverage
away from its starting level; (iii) lagged values of (M/B)_efwa retain explanatory power. The effect
has a half-life well over 10 years: capital structure in 2000 depends on market-to-book variation
from 1990 and earlier even after controlling for the 1999 value.
the lagged to contemporaneous coefficient, b₂/b₁ = 0.73); with 95% confidence at least two-thirds of
the (M/B)_efwa effect survives a decade. For **market leverage the effect is essentially 100%
permanent.** Historical market-to-book (even data over 10 years old) is more influential than current
market-to-book.
Why it matters
A foundational behavioral-corporate-finance result. It reframes capital structure as the long-lived
residue of market-timing attempts rather than the output of a static trade-off or a mechanical pecking
order, and it directly links market (in)efficiency to corporate financing. The (M/B)_efwa measure has
become a standard tool, and the paper anchors a large subsequent literature on whether managers exploit
mispricing and how durable those choices are.
Caveats
changing investment opportunities or risk is hard—market-to-book proxies for both.
Kayhan & Titman) argues firms do partially rebalance and that the long-run imprint is weaker or
reflects mechanical inertia in market leverage rather than active timing.
(M/B)_efwa exceeds 10.
Key references
Provenance: verified/generated from the paper's full text.