PAPER IS CONCERNED WITH:

1

• PAPER IS CONCERNED WITH
• Patterns of exchange rate volatility over time,
• Explanation(s) of Reduced Pass-Through,
• Trade Patterns
• Increased competition (pricing to market)
• Inflation Targets
• Policy recommendations

2
Patterns of exchange rate volatility over
timeEvidence (Figures 1-4) of increased
Canada-US nominal exchange rate volatility (after
the mid-1990s, but mostly after 2001). A test of
structural break?In the presence of reduced
pass-though, the resulting cost
isMicro-economic inefficiency that tends to
increase relative price variability (border
effects).
3
Measure of relative price variability

• SdtDev?(Pit/EtPjt) si,j
• Should show changes over time in this variable
  (add the mean of si,j in Table 1 for all
  sub-periods). The same dispersion index could
  be generated for the 21Canadian, 21 US and 49
  Canada-US pairs to see if price dispersion has
  increased over time similarly across and within
  borders.

4
Empirical evidence (Table 1 and Fig. 4)

• Distance and border positively affect si,j (as in
  Engel and Rogers (1996))
• Fig. 4 shows that the border effect coefficient
  increases dramatically between 2001 and 2002
  (September 2001?) How much of the increase in the
  border effect can be attributed to increased
  exchange rate volatility and reduced
  pass-through, and how much is due to some
  exogenous event (? transaction costs).
• Could remove this influence
• Regression of ?(EtPit/Pjt) on DUMMY2001 and
  construct a new si,j.

5
(2) Explanations of Reduced Pass-Through

• Monetary policy
• Taylor (2000) emphasizes the fact that the recent
  low inflation episode in the United States has
  also meant lower persistence of inflation.
  Therefore, lower pass-though is possible if there
  is a reduction in the persistence of cost and
  price changes (since expectations are changed).
• Consequently, the volatility in E is not
  transmitted to prices.

6

• Increased international competition
• To keep market shares, when there are substitutes
  available, firms will do some pricing-to-market ?
  reduced pass-through.
• Trade Patterns
• More intra-industry trading can lead to a smaller
  influence of exchange rate movements on prices.
• Recent increased demand for commodities (energy
  and non-energy) from Canada by emerging economies
  (mostly China), may have affected the exchange
  rate without affecting the local CPIs a lot.
• In all cases, a change in E will not affect P
  much, so the Central Bank (with an inflation
  target) will not change interest rates.

7
Policy

• Estimation of a Taylor Rule equation (Table 2)
  implies less weight on exchange rate after 1997,
  as should be the case in the presence of reduced
  pass-through and inflation targets.
• Evidence on the persistence of inflation in
  Canada should be presented to see if Taylors
  hypothesis that low and less persistent inflation
  is responsible for lower pass-through is
  validated over the sub-samples.

8
Can we conclude that the adoption of inflation
targets is responsible for reduced pass-through
and increased exchange rate volatility?

• The evidence is not conclusive yet.
• Low pass-through, low and stable inflation and
  high exchange rate volatility are compatible with
  several models (nominal wage rigidity, menu
  costs, pricing-to-market, noise traders,).
• A look at more disaggregated data might help shed
  some light on this issue.

9
Suggestions

• It would be nice to evaluate the speed at which
  exchange rate changes are incorporated into
  prices changes (p) at local levels.
• How?
• Regression of pi on lagged ?E (See Devereux,
  2000, BofC)
• Some patterns could be identified. Eg Bigger
  cities, with more competitive markets, might
  experience less pass-though than smaller ones
  (competition hypothesis).
• These regressions would be useful with data on
  different types of goods or individual goods
  (intra-industry hypothesis).

10
Concluding remarks

• This is a very nice paper that raises some
  important issues for the conduct of monetary
  policy.
• More work has to be done to determine the timing
  and the cause(s) of the observed increased
  exchange rate variability and reduce pass-through
  (how much is related to monetary policy and how
  much to other causes), but this is a promising
  first step.
• What can be said about which inflation rate (CPI,
  PPI, non traded, imports prices,..) to target?