David F. Hendry 7stimulates theory. Yet the latter is rarely the case in applied econometrics – and to
see how we arrived at such a state, we need to consider the contingent history of
our discipline.
1.3 Historical background
“The time has come,” the Walrus said, “To talk of many things: Of shoes- and ships – and sealing-wax – Of cabbages – and kings – And why the
sea is boiling hot – And whether pigs have wings.” (Lewis Carroll, 1899)
The histories of statistics and econometrics are now reasonably well documented:
on the former, see, e.g., the books by Stigler (1986, 1999) and Hald (1990, 1998);
and for the latter, see Epstein (1987), Morgan (1990), Qin (1993), Klein (1997),
and Le Gall (2007); also see Christ (1994), Spanos (2006), Farebrother (2006) and
Gilbert and Qin (2006); and for reprints of key papers, see Darnell (1994) and
Hendry and Morgan (1995), with related material in Caldwell (1993), Hamouda
and Rowley (1997), Mills (1999) and Campos, Ericsson and Hendry (2005). These
books provide overall bibliographic perspective.
1.3.1 Pre-Cowles
The shop seemed to be full of all manner of curious things – but the oddest
part of it all was, that whenever she looked hard at any shelf, to make out
exactly what it had on it, that particular shelf was always quite empty:
though the others round it were crowded as full as they could hold. (Lewis
Carroll, 1899)An aspect of that history which is still somewhat under-emphasized, despite being
stressed by Hendry and Morgan (1995), is the role that empirical studies have
played as a driver of new econometric concepts, theories and methods, stand-
ing in some contrast to its direct impact on economics. Certainly, early attempts
were replete with what we would now view as blunders – William Stanley Jevons’
sunspot, and Henry Moore’s Venus, theories of business cycles are regularly trotted
out as examples of how silly econometricians can be, yet Jevons (1875) and Moore
(1923) respectively need to be contrasted with other careful and insightful empir-
ical analyses in Jevons’ research, edited by Foxwell (1884), and in Moore (1911).
On the former, see the appraisal in Peart (2001); and on the latter, e.g., Stigler
(1962) comments: “Moore’s standard of craftsmanship is high: the basic data are
fully reported and the work was carefully done.” Also, note the cited comment
from Alfred Marshall to Moore in 1912 that “the ‘ceteris paribus’ clause – though
formally adequate seems to me impracticable,”a point that will recur below. The
“upward sloping demand curve” for pig-iron in Moore (1914) is perhaps the most
notorious misinterpretation, but in fact led to many later insights – in particu-
lar, reactions to it helped unravel the whole complicated and intertwined issues