Boudoukh 5 and 10 years horizons using

Boudoukh and Richardson (1993) argued that the negative
relationship between inflation and stock returns reported in existing studies
is due to the short-term asset returns with time horizons of less than one
year. They showed that using the annual instead of monthly or quarterly data,
to examine the relationship at long horizon (e.g. 5 years), a positive
relationship could be found, and results obtained from the two centuries data
of the US and the UK stock and bond markets empirically support their
supposition. Schotman and Schweitzer (2000) also compared the potential of
stocks against inflation for different investment horizons. They showed that
the negative inflation hedges potential stocks at the short run but can become
positive if the investment horizon changes to long run (over 15 years horizon),
which relies on inflation persistence: the higher inflation persistence the
better performance of stocks as a hedge against inflation. Engsted and
Tanggaard (2002) measured inflation and returns at 1, 5 and 10 years horizons
using a VAR model approach on the US and Danish stock and bond market. They found
a weak positive relationship in the UK stocks market at all three horizons
without showing an increase with a longer horizon, but a significant positive
relationship which becomes stronger as the horizon increases in Denmark stock
market, consistent with Boudoukh and Richardson (1993). Kim and In (2005) showed
a positive relationship between nominal stock returns and inflation at a long
scale (128-months period), but a negative one at most of the short scales (Lest
than 6-months period) in the US market from 1926 to 2000, which also supports
Boudoukh and Richardson (1993). Cardinale (2005) investigated the long-run
equilibrium relationship between stock market returns and consumer prices in a
cointegrating framework for the US, the UK, Germany and Japan and find mixed
results which are sensitive to the data horizon in choosing how many years of
lag. In addition, they find mixed support for the one-to-one equilibrium
hypothesis in different inflation regimes. Laopodis (2006) used the bivariate
and multivariate vector autoregressive cointegrating specifications to test the
dynamic interactions among the equity market, economic activity, inflation and monetary
policy under three monetary regimes and find a weak negative relationship between
the US stock returns and inflation during the period 1970s and 1980s, contrary
to previous cointegration studies. Pilotte (2003) also shows that the
relationship between inflation and stock returns varies across different
industries, but does not support the explanation given by Boudoukh et al.
(1994) for the variation being caused by economic fundamentals.