MACRO RESEARCH
May 2016
A short research note on inflation
Low inflation has been a concern for Western policymakers ever since the start of the Great Recession.
Despite record low interest rates and unprecedented balance sheet expansion most central banks are still
looking for higher inflation. With interest rates around zero and the economy stuck in a liquidity trap,
higher inflation is highly welcome as it reduces real interest rates and helps to erode elevated private and
public debt levels. In this research note we explore what the future might bring with regards to inflation.
In the first part we briefly look at the economic theory behind inflation. The second part then revisits two
important lessons from recent history. The third part builds on the first two parts and looks at what the
future might bring with regards to inflation. Part four summarizes all this in some concluding remarks.
1. Different types of inflation
In essence, we can distinguish between three types of inflation: (1) cost-push inflation, (2) demand-pull
inflation, (3) monetary inflation. This is also the framework we use when we internally discuss the outlook
on inflation.
1.
Cost-push inflation
Cost-push inflation is mostly the result of higher commodity prices or an acceleration in wage growth.
Inflation is pushed higher by the increase in the price of production inputs even when demand remains
stable. Examples include the negative oil-supply shocks and the following wage-price spiral in the 1970s, a
period charachterized as stagflation.
2.
Demand-pull inflation
Demand-pull inflation is driven by stronger demand for goods and services among consumers and
investors. Prices increase as more money chases after a limited available amount of products. The above
target inflation rates witnessed in several Southern European economies prior to the Great Recession, for
example, fall into this category.
3.
Monetary inflation
The monetary theory of inflation holds that increasing the money supply leads to higher inflation. The
concept is mostly associated with Milton Friedman who said that “inflation is always and everywhere a
monetary phenomenon”. According to this view, inflation has little to do with things like production costs or
consumer demand but everything with the supply of money.
2. Two important inflation lessons from the recent past
For those of us who had used the right economic model during the previous eigth years (i.e since central
banks started expanding their balance sheets through measures of quantitative easing), inflation has
behaved pretty much as expected. The presence of huge slack in the labor market (high unemployment) and
low capacity utilization rates in the manufacturing sector for example implied that inflation has been
experiencing downward pressure.
Macro Research Contact
Hans Bevers | Chief Economist |+ 32 2 287 97 04 | [email protected]
1|6
Two important lessons from the past 2008 experience with regards to inflation are best kept in mind. The
first one is that huge expansions in the size of the central bank balance sheet (the monetary base) do not lead
to the same increase in the broader money supply (M2 or M3) nor result in runaway inflation under liquidity
trap conditions. The Japanese example at the beginning of the millenium already made this point very clear
and the same has been witnessed in recent years (see graphs below). In other words, the inflationistas were
proven wrong.
The Japanese example already suggested that QE would not lead to runaway inflation after 2008
200
Japan
Monetary base
M2
CPI index
180
600
500
160
400
140
300
120
200
100
100
80
Start 2001 = 100
2001
2002
2003
Start 2008 = 100
0
2004
2005
2006
2007
United States
Monetary base
M2
CPI index
2008 2009 2010 2011 2012 2013 2014 2015 2016
A second lesson is that those who predicted outright deflation (the deflationistas) were also proven wrong.
Indeed, the experiences from previous decades would have predicted a very outspoken clockwise
disinflation spiral in line with the so-called accelerationist Phillips-curve (see right graph below). Most
economists now agree that this can be explained by both better anchored expectations with regards to
inflation and downward nominal wage rigidity (i.e. an important element of sticky wages even with the
presence of significant slack in the labor market).
No outright deflation after 2008 even though the Phillips-curve would have predicted otherwise
3
11
2.5
2
1980s
9
1990s
8
2000s
7
6
1.5
5
4
1
3
2
0.5
1
Core inflation in G7 countries
Unemployment rate
0
0
Q2 1996
1970s
Core inflation United States
10
Q2 2001
Q2 2006
Q2 2011
Q2 2016
3
4
5
6
7
8
9
10
11
3. What’s next for inflation?
The answer to this question is important yet very uncertain at the same time. After all, making economic
predictions is difficult, certainly about the future.
In the short run, despite the general very low level of commodity prices, base effects will send headline
2
inflation higher in late 2016 as the earlier sharp drop in energy and metal prices disappears from the year on
year comparison. This evolution is clearly shown in the left graph below. But as you can see, the headline
inflation rate is very vulnerable to sudden changes in commodity prices. That’s why central bankers and
economists tend to focus more on the underlying pressures of inflation.
Headline inflation set to pick up throughout 2016
4
70
50
3
30
2
Underlying inflation still modest for now
4
3
2
10
1
-10
1
-30
0
G7 headline inflation (lhs)
Global commodity prices (YoY, rhs)
-1
Q2 1996
Q2 2001
Q2 2006
Q2 2011
0
-50
-1
-70
-2
Q2 2016
UK
US
Euro area
Japan
2006
2008
Core inflation
2010
2012
2014
2016
Core inflation (which strips out the the more volatile elements of food and energy) looks set to remain fairly
modest for now. Moreover, differences between regions could exist. In Japan and the eurozone for example,
demand is still fairly weak suggesting that inflation remains subdued and that the BoJ and ECB are facing a
very difficult challenge to meet their 2% inflation target anytime soon. The situation in the US is a bit more
complex. Core inflation has been strenghtening a bit in recent months in line with the improvement seen in
the labor market. That said, at 1.6%, the Fed’s most preferred measure of core inflation remains fairly
modest. Accelerating wage growth (from low levels) and stalling productivity growth signal that unit labor
costs are picking up. In case wage growth would accelerate further and the productivity slump would
continue (which still needs to be seen), then core inflationary pressures would obviously increase. Surveys
measuring the underlying price evolution, on the other hand, signal that core inflation remains modest for
now.
Labor market slack weiging on eurozone inflation
Weakening inflation (expectations) in Japan
4
110
3
Core CPI
3
90
2.5
1
70
2
0
50
1.5
30
1
10
0.5
2
-1
-2
Core inflation
Headline Inflation
Household inflation expectations (rhs)
-3
-4
2004
2006
2008
2010
2012
2014
Unemployment rate
-10
2016
0
6
7
8
9
10
11
12
13
That said, the outlook for inflation in a medium to longer-term perspective is subject to more uncertainty.
One should always bear in mind that the future could look different. Keynesians for example went into the
1970s thinking that inflation and weak economic activity were always mutually exclusive (in line with the
Phillips curve described higher). But stagflation was real in the 1970s just as Milton Friedman predicted.
3
Indeed, the combination of a huge adverse supply-shock (sharp rise in oil price) in combination with (too)
loose monetary policy basically led to entrenched expectations of higher future inflation. Therefore, it’s
impossible to exclude this as a possible scenario going forward (see here or here for example). Of course,
once Paul Volcker started to raise rates to kill off inflation (expectations), the Phillips curve was shown well
alive and kicking again and this has basically been the case ever since (except for the accelerationist element
more recently as described in the first part of this note but this is of secondary importance). Therefore we
cannot rule out a scenario of stagflation for example.
At the onset of the Great Recession, many observers warned that central bank liquidity injections would
rapidly lead to soaring inflation. In the long run there is indeed a strong relationship between money growth
and inflation (see graph left below). Obviously, as stated higher, the Japanese example and the post-2008
experience have shown that conditions are difficult in a liquidity trap. First of all, in case of the latter, as we
have seen, the monetary base and the money supply (M2 or M3) behave differently. The monetary base has
been increasing far more strongly than the money supply due to a dysfunctioning of the credit transmission
mechansim against the back of a highly leveraged private sector.
Secondly, the velocity of money (or the frequency at wich money changes hands) should be taken into
consideration. According to the monetary theory of inflation: M (money supply) x V (velocity of money) = P
(general price level) x T (number of transactions). What follows is that, in order for the money supply to be a
real key driver of inflation, one should more or less assume that V and T remain constant. In reality,
however, the velocity of money has been declining rapidly ever since 2008 (see right graph below).
LT-relationship between money growth and inflation
Rapid decrease in velocity after 2008
2.5
Inflation (1960-2000)
2.3
2.0
1.8
1.5
1.3
Money growth (1960-2000)
Velocity of M2, United States
1.0
1966
1976
1986
1996
2006
2016
4. Concluding remarks
Both the inflationistas and the deflationistas have been proven wrong by events since 2008. Runaway
inflation has not materialized. In fact, earlier liquidity trap evidence from Japan would have predicted just
that. Outright deflation did not occur either which was a bit surprising for those who would have thought
that the Phillips-curve would behave in line with earlier decades.
Despite record low interest rates and unprecedented balance sheet expansion most central banks are still
looking for higher inflation. Almost 8 years after the start of the Great Recession, global demand is still fairly
weak implying that underlying inflationary pressures should remain modest in the Western world and
certainly in Japan and Europe. Extra budgetary stimulus could help monetary stimlus to gain traction.
In the US, the combination of accelerating wage growth (although still at a modest level) and stalling
productivity growth suggest that core inflation should move to around 2%, in line with the Fed’s target.
4
Surveys measuring price levels (a good leading indicator in historical context), on the other hand, indicate
that inflation remains modest.
But even if inflation exceeds the Fed’s target somewhat, this will not be a major source of concern for
monetary policymakers given that they missed their target for many years. Markets might need to revise
their expectations upward as they are very dovish with regards to the future path of rate hikes (see graph
below). But this is not new and, admittedly, until now they have done a far better policy rate forecast job
than for example the Fed’s own board members.
Liquidity trap economics
8
7
Fed
ECB
6
BoE
5
BoJ
4
3
2
Market implied policy rate
1
0
Central bank policy rate
-1
Q2 1991 Q2 1994 Q2 1997 Q2 2000 Q2 2003 Q2 2006 Q2 2009 Q2 2012 Q2 2015 Q2 2018
Obviously, in a longer term perspective, the outlook is much more uncertain against the background of
unprecedented money printing. Once the private sector debt deleveraging process has run its course, the
transmission mechanism is fully restored and consumers and enterprises really would start to spend again,
then inflation could rapidly move higher in line with the monetary theory of inflation. An if history is a
guide, we cannot even rule out a scenario of stagflation.
But worsening demographics should also be taken into account. Both the BIS (2015) and Morgan Stanley
(2015) find that population ageing could lead to upward wage and inflation pressures. The decrease in the
relative share of the working age population, in this view, will make labor more scarce and therefore more
expensive pushing inflation higher. The IMF (2015), on the other hand, finds a positive relationship between
ageing and disinflation through their influence on lower aggregate demand and slow supply responses.
Moreover, future technological developments will also play a role. It’s not clear at all how fast the process of
automation and disruptive technological progress will proceed, let alone what the effects will be on prices.
The early 1980s have shown that central banks can kill off undesired inflation quite easily (albeit at the cost
of recession). Recent evidence from Japan, the Eurozone and the US has shown that it is more difficult to
prop up inflation when it is too low. With interest rates near the zero lower bound, this can be a real
problem. Liquidity trap conditions might persist for some time. This may also mean that central banks will
need to look for bolder instruments (for example by introducing a higher inflation target or helicopter
money) to fight off the threat of deflation. The future remains uncertain indeed.
Hans Bevers
Chief Economist
5
References
C. Goodhart, Pradhan, M., Could Demographics Reverse Three Multi-Decade Trends?, Global Issues,
Morgan Stanly, September 2015
Y. Yoon, Kim, J., Impact of Demographic Changes on Inflation and the Macroeconomy, IMF Working Paper,
November 2014
The information contained in this document is provided for pure information purposes only. Present document does not constitute an investment advice
and independent investigations, assessments or analysis regarding any investment should be undertaken by the potential investors and recipients as
deemed appropriate by them. This document doesn’t form part of an offer or solicitation for shares, bonds or mutual funds, or an invitation to buy or sell
the products or instruments referred to herein. Applications to invest in any fund referred to in this document can only validly be made on the basis of the
current prospectus or simplified prospectus, together with the latest available annual report and accounts. All opinions and financial estimates herein
reflect a situation on the date of issuance of the document and are subject to change without notice. Indeed, past performances are not necessarily a guide
to future performances and may not be repeated. Degroof Petercam S.A. has made its best efforts in the preparation of this document. The information is
based on sources which Degroof Petercam S.A.believes to be reliable. However, it does not represent that the information is accurate and complete.
Degroof Petercam S.A. is acting in the best interests of its clients, without carrying any obligation to achieve any result or performance whatsoever. Degroof
Petercam S.A., its connected persons, officers and employees do not accept any liability for any direct, indirect or consequential loss, cost or expense arising
from any use of the information and its content. Present document may not be duplicated, in whole or in part, or distributed to other persons without prior
written consent of Degroof Petercam S.A.
6