Special Macro Comment Inflated Fears

13 May 2009

Special macro comment Serdar Küçükakın, senior economist

Inflated fears •

Inflation fears are somewhat overstated

•

Central banks have plenty exit strategies

Will the extraordinary expansion of central bank balance

very negative throughout 2010 and beyond, with growth

sheets lead to a serious bout of inflation? The short answer to

returning, but remaining at or below trend. In other words, real

this question is most probably not. In this article we outline the

factors will provide a strong disinflationary force for some time

main justifications for this view. Firstly we present a simple

to come.

framework to suggest that market participants should focus on expectations rather than “real” factors in seeking to understand

World economy: output gap

inflation risks in the coming years. In particular we examine

%

how inflation expectations could derail even if the real economy is in the doldrums. Secondly we show how to manage the different variants of exit strategies. The Phillips curve To understand why current inflation risks are low, it is important to look beyond slogans and take a broad look at the economics of inflation. In the standard expectations-augmented Phillips curve, inflation can arise because of two events: an overheating economy or because inflation expectations lose their anchor:

4 2 0 -2 -4 -6 1980

85 World

90 95 2000 5 10 14 Advanced economies Emerging economies

Source: IMF

П = Пe + a x (Y – Yp) Even if we disregard the output gap and its effects on inflation where Π is inflation, Πe is expected inflation and Y – Yp is the

there is other powerful evidence to suggest that a fragile

output gap (output minus the potential level of output).

recovery will not lead to inflation.

According to this theoretical framework, policy changes cause inflation either by stimulating spending, thereby pushing the

One of the most famous dictums of economics is that inflation

GDP above its potential, or by causing the public to raise its

is ultimately always a monetary phenomenon. Economic

inflation expectations and thus raise prices in anticipation of

textbooks describe the process as starting with cash and

inflation.

reserves in the banking system (the “monetary base”). If the central bank increases the monetary base in a normally

Output gap

functioning banking system, there will be a multiplier effect on

If we return to reality, what does this mean for the future

loans. Banks lend out the excess cash they have, and in turn

inflation outlook? As the following diagram shows, the IMF’s

these loans are spent and redeposited at banks. This in turn

calculations up to 2014 indicate that output gaps do not pose a

creates new bank lending power, new deposits and so on. In

serious inflation risk. The world output gap turned negative in

the US, for example, the money multiplier was around 9 up to

2008, and so far this negative output gap has only got bigger,

the beginning of last year. In other words, USD 1 of monetary

meaning that the difference between the potential output and

base supported around USD 9 in bank loans. This surge in

the real output is growing. In fact the IMF’s calculations

loans in turn increases nominal purchasing power, and this

suggest deflation and not inflation, especially in 2010. More

eventually results in a surge in prices. Today’s complex credit

importantly, the general expectation of a feeble recovery

markets admittedly complicate the process, but this is the core

around the world implies that the global output gap will remain

idea behind the famous dictum.

SERDAR KÜÇÜKAKΙN +31 (0)20 629 5086

ECONOMICS DEPARTMENT

13 May 2009 US: monetary base and money multiplier

zone and Japan, will be fragile (partly because of the disfunctioning of the credit market), there is little danger that credit growth will lead to a rise in inflation expectations.

120

10

100

9

80

8

Inflation expectations

60

7

How then can monetary policy today generate high inflation

40

6

tomorrow? The answer lies on the first term of the Phillips

20

5

curve: inflation expectations. Inflation can increase simply

0

4

because people expect inflation and so adjust their wages and

3

prices accordingly to avoid being surprised by rising inflation.

-20 00

01

02

03

04

Monetary base (%yoy, lhs)

05

06

07

08

09

M2 money (multliplier, rhs)

Source: Thomson Financial, calculations ABN AMRO Econ. Department.

There are (theoretically) three relevant ways of loosening the grip on inflation expectations in today’s economic environment. Firstly, as can be witnessed around the world, high levels of unemployment can prompt central banks to ease aggressively.

So why don’t we believe this story? Firstly, in “normal” times

However, systematic attempts to exploit the relationship behind

this enormous increase in the size of central bank balance

the inflation-output trade-off can backfire. A prolonged period

sheets would indeed lead to a huge jump in consumer

of easy money could result in businesses raising their inflation

purchasing power, with inflation as a result. However, there is

forecasts rather than hiring more people.

nothing normal about today’s market: banks are not lending and the surge in the monetary base is consequently piling up in

The second risk of rising inflation expectations comes from the

excess reserves. The above diagram shows that while the

interaction between the budget deficit and monetary policy.

monetary base in the US has skyrocketed, the US money

The US and other governments face major deficit and debt

multiplier has collapsed.

challenges in the coming years. Normally an independent central bank would not be willing to monetize debt. In the

Secondly, historical evidence suggests that recoveries from

current economic setting, however, there is growing political

recessions are “creditless”. If we zoom in on the US, we can

pressure on central banks to offer a helping hand. In the US,

see that in the post-Second World War period (1951 - 2008)

Bernanke’s four-year term as Fed chairman ends in January

there was no correlation between bank lending to businesses

2010, and a populist chairman might be tempted to reduce the

and growth of GDP (0.22). In fact, the correlation between the

real value of debt by creating surprise inflation.

two variables increases considerably if we lag credit growth by three quarters (0.50).

Government debt % yoy

However correlation does not necessarily imply any causation. To overcome this shortcoming we have applied the Granger

25

causality1 test to see whether there is any causality between

20

GDP and lending. This test suggests that a value of GDP in T

15

gives significant information about the value of lending in T+1. The outcome of these two econometric tools could at first sight be counterintuitive because growth in credit could be expected to lead to economic expansion. The reality would seem to be that businesses only start borrowing when they see clear signs of recovery in the economy, or the flip side of the coin: banks only become less cautious about lending when there are clear

10 5 0 US

Eurozone 2008

Japan 2009

UK

2010

Source: Thomson financial, forecasts ABN AMRO Economics Department.

signs of recovery. Even without politicizing the central bank, debt dynamics could Taking these observations as our starting point and also taking

force the Fed’s hand, as central banks, preferring the lesser of

into account that any recovery over the coming years, not only

two evils, may choose to monetize debt in order to avoid an

in the US but also in other major economies such as the euro

implicit default. This risk naturally rises as the debt burden

1

The Granger causality test is a technique for determining whether one time series is useful in forecasting another. A time series X is said to Grangercause Y if it can be shown, usually through a series of F-tests on lagged values of X (and with lagged values of Y also known), that those X values provide statistically significant information about future values of Y.

SERDAR KUCUKAKIN +31 (0)20 629 5086

becomes higher. The huge increases in debt/GDP ratios over the next few years means this is a real concern, especially in a scenario of prolonged weak growth.

ECONOMICS DEPARTMENT

13 May 2009 The final risk relates to the self-fulfilling nature of the need for a

once it gets anchored in people’s expectations, is more difficult

stable store of value. Currencies are used as a store of value

than battling inflation. Japan’s experience in the 1990s is the

in a low-inflation environment. However the risk of higher

biggest proof of this. Inflation rates across the globe have

inflation may lead investors into other assets that hedge

come down significantly in recent months because of falling

inflation better. In other words, a prolonged period of ultra-easy

consumer demand and commodity prices.

monetary policy (= quantitative easing) can call into question a central bank’s credibility and result in a move away from

US: inflation

“nominal” assets, a dynamic that can itself fuel inflation.

% yoy

If we look at the three (theoretical) possibilities described above that could lead to rising inflation expectations, we believe that the second and third could pose a threat further down the road. The rationale behind this thesis lies in the expectation of a somewhat sluggish recovery of the economic cycle. Managing the exit strategy from quantitative easing (QE) and choosing the right exit tool and timing will be critical (the day after). We elaborate on this below.

14 12 10 8 6 4 2 0 -2

8 7 6 5 4 3 2 1 0 80

88

92

96

00

Expected inflation rate in 5 years (lhs)

The first option – high and still rising unemployment levels that could tempt central banks to loosen more aggressively – has

84

04

08

Actual inflation (rhs)

Source: Thomson Financial

little relevance on a global scale because major central banks such as the Fed, the Bank of Japan and the Bank of England

Zooming in on the US, as shown in the above graph, we can

have already slashed their policy rates close to zero and are

indeed

applying QE to battle the adverse effects of the credit crisis.

considerably higher than actual inflation. However this does

Furthermore, although lagging somewhat behind, the ECB has

not mean that inflation expectations have recently risen

also already lowered its policy rate by 325 basis points to

significantly. On the contrary, the current positive difference

1.00% since July 2008. Furthermore, ECB president Jean-

between expected inflation and actual inflation is attributable

Claude Trichet announced after the latest rate cut in May that

solely to the fact that the latter has dropped sharply owing to

the ECB would start 12-month refinancing operations. So far,

falling commodity prices and retreating demand. Inflation

the ECB’s refinancing operations have had a maximum

expectations have actually been hovering between 3% and 4%

maturity of six months. Trichet also announced that the ECB

since the beginning of the 1990s. In other words, the Fed has

would start building up a portfolio of covered bonds in order to

been quite successful in anchoring inflation expectations for

improve the functioning of this market.

almost twenty years. This conclusion is very important because

see

that

inflation

expectations

are

currently

it shows that the American public has great trust in its central

Inflation

bank and therefore, despite big swings in actual inflation, does not easily change its inflation expectations.

% yoy

6 5

How to un-QE?

4

While the markets have interpreted the current actions by the central banks as highly necessary in order to counter the very

3

sharp economic downturn and the continuing disfunctioning of

2

the financial system and although the current inflation

1

expectations seem well anchored, the public mood could

0

change further down the road. At that juncture it will be

-1 00

01

02

03 US

04 05 Eurozone

06

07 UK

08

09

Source: Bloomberg

essential for central banks to demonstrate their utmost skills to undo the current QE. In this respect the speed at which central banks

can

unwind

their

current

positions

is

key

to

understanding the risks of inflation. We see three alternative exit strategies:

Market participants across the world have generally interpreted these actions of the major central banks as being necessary. In

1. Passive style: A passive strategy in which central bank

the current setting, with a synchronized global downturn, the

balance sheets gradually shrink as short-term assets

central banks are actually very keen to create some kind of

mature and reserves are “returned”.

inflation or expectation of inflation because battling deflation, SERDAR KUCUKAKIN +31 (0)20 629 5086

ECONOMICS DEPARTMENT

13 May 2009 2. Active style: If the passive strategy is too slow, central banks can sell the credit and other assets they have accumulated, thus drawing reserves out of the system. 3. Issuing debt: Central banks can issue their own debt (or have the Treasury issue extra debt and deposit it at the central bank). This extracts reserves from the banking system, thus preventing an expansion in loans and money, as central bank debt has a lower “multiplying” effect than reserves. While theoretically central banks have plenty of “exit” tools at their disposal, the reality is more challenging because the Fed and the BoE, for example, are rapidly building up all kinds of assets, thus also longer-term and less liquid assets. This means that a passive exit strategy (alternative 1) would take a relatively long time to complete, and that is time that a central bank may not have if inflation expectations start rising. Any exit strategy is therefore likely to involve some elements of active strategy. In particular, the faster the banks and credit markets return to their traditional role, the more likely it is that central banks will have to participate actively in the unwinding of QE (alternatives 2 and/or 3). In short, we believe that the risks of high inflation in the current setting and further down the road are somewhat overstated. Firstly, as shown for the US, the public does not change its inflation expectations that easily. In other words, central banks’ past track record with regard to inflation plays an important role in determining inflation expectations. Secondly, even if the public mood changes further down the road, central banks have plenty of exit strategies from the current QE. However, it must also be mentioned that a strong economic recovery will make it easier for central banks to undo their recent activities. In the event of a fragile recovery central banks will have to use all their steering skills to unwind their balance sheets in an orderly fashion because they will not want to cap any incipient recovery. Furthermore central banks could also face pressure from politicians and special interest groups lobbying against a quick exit from extraordinary credit policies. The weaker the economic recovery, the stronger these pressures are likely to be.

Important information The views and opinions expressed above are subject to change at any time. Individuals are advised to seek professional guidance prior to making any investments. This material is provided to you for information purposes only and should not be construed as advice or as an invitation or offer to buy or sell securities or other financial instruments. Before investing in any product of ABN AMRO Bank N.V., you should obtain information on the various financial and other risks and the possible restrictions that you and your investment activities may encounter under applicable laws and regulations. If, after reading the brochure, you consider investing in this product, you are advised to discuss such an investment with your relationship manager or personal advisor and check whether this product – considering the risks involved – accords with your investment activities. The value of your investments may fluctuate. Past performance is no guarantee for future returns. ABN AMRO Bank N.V. has taken all reasonable care to ensure that the information contained in this document is correct, but does not accept liability for any misprints. ABN AMRO Bank N.V. reserves the right to make amendments to this material.

SERDAR KUCUKAKIN +31 (0)20 629 5086

ECONOMICS DEPARTMENT