FEDERAL RESERVE BANK OF ST. LOUIS
WHAT WENT WRONG AND WHEN?
‘Two fundamental r’elationships between Ml and specific economic measures have been suppom’ted em— pim’icallv for’ decades. One r’elationship is the link be- tween mnoney and GNP, a measur-e of total income in the economy. The second relationship is the link be- tween money and pr’ices. Charts 1 and 2 show the dramatic changes in these relationships that occurred during the 1980s.
Chart 1 depicts the behavior of the income velocity GNP divided by Ni 1 I for the past 40 years; as the chart
unusual occurred to velocity
around 1982. From 1946 through 1981, it rose fairly steadih’ at about 3.6 percent pervear; since then, it has
declined at an annual rate of about 2.4 percent.
Chart 2 shows Ihe r’elationship since 1948 between annual inflation (as measured by the growth of the
GNP deflator’) and the average growth in Nil over a
three—year’ per’iod; use of Ml’s trend growth is de- signed to capture the long-run impact of money on
prices. While the rate of inflation deviated from the trend growth of Ml, sometimes substantially, from
1948 to 1981, the deviations generally wer’e ternporan’. More importantly, the lar’ger deviations were attritiut—
able to non-monetary events
for example, govern—
nient mandated wage—price controls, OPEC oil pr’ice actions and the like;. Since 1982, however’, inflation has
been substantially and per’sistentlv below the trend
growth in Ml. ‘I’hese deviations able to a specific non—monetai
are not event.
Numerous attempts have been made to explain the r’ecent changes in velocity. In this paper’, these expla- nations are gmnuped loosely into three categor-ies: mis- specification, a portmanteau categon’ we call ‘‘str-uc— tural shifts’’ and cyclical factors.’
The most widely used velocity measure, the income velocity of Ml, is calculated by dividing nominal GNP by the nominal stock of Ml. Both GNP and Nil are empirical counterpar1s to theoretical concepts that appear’ in various theories of the demand for money. One explanation for’ the shift in velocity is that GNP or
‘.A number of these are considered in studies by Rasche (1986), Darby et, al, (1987). Hetzel (1987), Trehan and Walsh (1987) and Kretzmer and Porter (1987). The categories considered here are somewhat more general than those considered by Trehan and Walsh.
Nil or’ tioth have become less reliable pr’oxies for their- cor’r’esponding theor’etical concepts. This protilem is called a specification problem
GIVP Vs. Transactions Measures
One specification problem could arise if money is field pr’irnam’ilv to make daily ti’ansactions.’ if these include intermediate and financial tr’ansactions. the usual velocity measure could vary with changes in the propor’tion of such transactions relative to transac—
tions on final goods and services. Because GNP mea-
level of expenditur’es on all tran.saclrorrs, In this case, GNP is a useful proxy for total tr’ansactions only if the propor’tion of GNP to total tr’ansactions remains rela— tiyely constant.
This protilem can manifest itself in sever’al ways. F’or’
example, suppose consumers purchase mor-e goods and, as a result, imIcr-ease their money holdings in
proportion to their incr’eased desire to spend. If these
newl purchased goods domestic inventories of
are impor’ted or’ dr’awn from preyiriusly produced goods,
GNP will remain unchanged while the demand for’
money rises. Consequentl~ the usual measur’e of ve- locity would decline, while an alternative measure based on total transactions would r’emain unchanged.
Thus, using GNti as the transactions measure to calcu- late velocity may pr-ociuce sizable swings in velocit~’
whenever ther’e are large swings in inventories or’ net exprir’ts. Some analysts have argued that gr’oss domes-
tic final demand IGDFDI, which equals GNP TTdJIUS inventory adjustments and net exports, is preferable
to GNP as the transactions proxy.’ t,Jnforftrnately, the substitution of GDFD for- GNP does not explain the
velocity ptlzzle of the l980s. As chart 3 indicates, this velocity measure performs essentially the same as the
usual mneasur’e both before arid after 1981. Conse-
quently, siriiply replacing GNP with GDFD does not explain the protracted velocity decline dur’ing the
‘See the appendix to Thornton (1983) for an illustration of the specification problem involved in tinding the appropriate measure of
“income.” ‘There are Iwo distinct, though not mutually exclusive theories of the
demand for money: the transactions approach and the asset ap- proach. The asset approach emphasizes the role of money as an
asset and, hence, as an alternative way of holding wealth. The transactions approach emphasizes the role of money as a medium of exchange. For a useful discussion of this distinction in relation to the velocity issue, see Spindt (1985).
‘Radecki and Wenninger (1985). 6flasche (1986) also relects this explanation for much the same reason.