Lecture 9: Intermediate macroeconomics, autumn 2007
Lars Calmfors
1
Theory of consumption
Keynesian consumption function
C = C(Y – T)
Consumption depends on current disposable income
0 < MPC < 1
• But it is more reasonable to believe that consumption
depends on forward-looking decisions: Irving Fisher,
Milton Friedman, Franco Modigliani and Robert Hall
• Intertemporal decisions
• Fisher’s two period model
2
Figure 16.1 The Keynesian Consumption Function
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
3
Intertemporal budget constraint
Period 1:
S = Y1 - C1
Period 2:
C2 = (1 + r)S + Y2
Substitution of (1) into (2) gives:
C2 = (1 + r)(Y1 - C1) + Y2
C1 = 0 ⇒ C2 = (1 + r) Y1 + Y2
C2 = 0 ⇒ C1 = Y1 + Y2 /(1 + r)
C1 = Y1 and C2 = Y2 is always possible
C1 + C2 /(1 + r) = Y1 + Y2 /(1 + r)
(1 + r) is the price of consumption in period 1 in terms of lower
consumption in period 2. It is thus always more expensive to
consume in period 1 than in period 2.
r = 0 ⇒ C1 + C2 = Y1 + Y2
Present value of consumption = Present value of income.
4
Figure 16.3 The Consumer’s Budget Constraint
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
5
Figure 16.4 The Consumer’s Preferences
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
6
Figure 16.5 The Consumer’s Optimum
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
7
Figure 16.6 An Increase in Income
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
8
Figure 16.7 An Increase in the Interest Rate
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
9
• Expected future income changes influence consumption
already now
- Oil revenues in Norway
- Future pensions
- Anticipated future productivity increases in the US:
explanation of low savings and large current account
deficits
• Consumption smoothing
• Households try to smooth consumption over time (equalise
marginal utility of consumption)
- decreasing marginal utility of consumption
- the same consumption level each period if subjective
discount rate = market interest rate
10
Figure 16.8 A Borrowing Constraint
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
11
Figure 16.9 The Consumer’s Optimum With a Borrowing Constraint
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
Borrowing constraints
• Around ¼ of households are rationed in the credit market
• The MPC of rationed households is unity (one)
• A temporary income increase of ΔY gives a permanent
income rise by rΔY (the permanent return if the income rise
in invested in the credit market) for non-rationed
households. MPC ≈ r
• Hence, aggregate MPC = ¼ x 1 + ¾ x r ≈ 1/4
If consumption in each period depends on expected life income,
it should follow a “random walk”, that is consumption changes
cannot be predicted: they occur only when there are unexpected
news (as with stock prices and exchange rates) – Robert Hall
Ct = α0 + α1Ct-1 + εt
Time-inconsistent preferences
• Behavioural economics
• Too low savings because of ”pull of instant gratification”?
Question 1: 1000 SEK today (A) or 1100 SEK tomorrow (B)?
Question 2: 1000 SEK in 100 days (A) or 1100 SEK in 101 days (B)?
• Many people choose A in question 1 and B in question 2.
• This is an example of time inconsistent preferences.
• Individuals do not adhere to a long-term plan but deviate
from it.
Franco Modigliani’s life cycle hypothesis
R = Remaining years of work
Y = Annual income
W = Wealth
T = Remaining years of life
C = (W + RY)/T
C = W/T + RY/T
T = 50, R = 30 ⇒ C = W/50 + 30/50Y = 0,02W + 0,6Y
MPCW = 0,02
MPCY = 0,6
T = 21, R = 1 ⇒ C = W/21 + 1/21Y≈ 0,05W + 0,05Y
Figure 16.10 The Life-Cycle Consumption Function
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
Figure 16.11 How Changes in Wealth Shift the Consumption Function
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
Figure 16.12 Consumption, Income, and Wealth Over the Life Cycle
Mankiw: Macroeconomics, Sixth Edition
Copyright © 2007 by Worth Publishers
• Changes in asset prices (shares, houses) nowadays play a large
role for the development of private consumption
• Risks of “boom-bust cycles” – sudden “asset price reversals” tend
to reinforce cyclical variations
- property price bubble in Sweden, Finland and the UK in the 1980s
and “asset price deflation” in the early 1990s
- similar developments in Japan in the 1980s, after that prolonged
recession (depression)
- worldwide boom in stock prices in the late 1990s, then stock price
falls when the dotcom bubble burst
- are we now watching the beginning of a prolonged fall in house
prices (US, UK, Ireland, Spain, Sweden????)
• Difficult problem for central banks: Should they just have inflation
targets for the CPI or should they also try to counteract large swings
in asset prices (as Alan Greenspan and the Fed have done several
times before)?
- if asset prices rise too much, they may later fall a lot and make it
impossible to avoid a deep recession and deflation (since the
nominal interest rate cannot become negative: Japan is a prime
example)
- are central banks better than financial markets in identifying
asset price bubbles?
- Riksbanken has been criticised for an unclear policy strategy
- ECB uses money supply increases as an indicator of the risks of
asset price bubbles
Fig. 2.1
Real house prices
200
Index, 2000=100
200
Spain
180
180
160
160
Italy
Ireland
140
140
120
120
Euro area
100
100
80
80
2000
2001
2002
2003
2004
2005
2006
Source: OECD (2006).
EEAG Report 2007
The monetary situation in the euro area
Fig. 1.13
1)
12
Changes in M3 money supply
%
Over previous month
10
%
12
2)
10
8
8
6
6
4
4
ECB
reference value
Over previous year
2
2
0
0
1999
2000
2001
2002
2003
2004
2005
2006
3)
14
Credit growth
%
%
Lending for house purchase
12
14
12
10
10
8
8
6
6
4
2
4
Corporate credit
Consumer credit
2
0
0
1999
2000
2001
2002
2003
2004
2005
2006
1) M3=Currency in circulation, overnight deposits, deposits with agreed maturity up to two years;
index, seasonally adjusted, 3-month moving average in % (centred).- 2) Extrapolated to annual rates.
3) Changes over previous quarter; corporate credit = credit to nonfinancial corporations; Lending
to house purchase = credit to private households and NPISH.
Sources: European Central Bank; Ifo Institute calculations.
EEAG Report 2007
10
Savings and the pension system
• Pay-as-you-go system (fördelningssystem) – each generation
pays the pensions of the previous generation
• A funded system (premiereservsystem) – each generation
pays for its own pensions through pre-funding, which gives
a higher savings rate
• If one introduces a pay-as-you-go system, the first
generation in the system is a winner (since it does not pay
for any pensions for the preceding generation): our earlier
ATP-system, which was introduced in the 1960s
• The earlier ATP-system was not sustainable: it built on too
optimistic projections of future growth: this was the
background of the Swedish pension reform in the 1990s
• Problem: if a pay-as-you-go system is replaced by a funded
system, the last generation in the pay-as-you-go system
becomes a loser (one has to pay twice: first for the pensions of
the previous generation and then for the own pensions)
• Swedish pension reform: combination of a pay-as-you-go
system (the larger part is an actuarial pay-as-you-go system
where all labour income earns pension rights) and a funded
system (the PPM system)
• The Swedish pay-as-you-go system is based on defined
contributions and not as before on defined benefits
- benefits are automatically adjusted to contributions
- benefits are indexed to the developments of wages per
employed
- automatic brake adjusts benefits downwards if the
financial viability of the system is at risk
• Many other countries would need to do similar pension
reforms as in Sweden
- higher contributions
- lower pensions (Finland: indexation to average longevity)
- higher retirement age (Denmark: indexation to average
longevity)
- partial shift to funded system
The effects of ageing on per capita output
Dependency ratio
Austria
Belgium
Denmark
Finland
France
Germany
Greece
Ireland
Italy
Luxembourg
Netherlands
Portugal
Spain
Sweden
UK
EU 15-Average
2004
0.24
0.27
0.23
0.23
0.25
0.27
0.28
0.17
0.29
0.21
0.21
0.24
0.25
0.27
0.24
0.24
2050
0.55
0.47
0.42
0.46
0.46
0.49
0.62
0.41
0.65
0.35
0.42
0.53
0.68
0.47
0.38
0.49
Participatio
n rate
80.00
67.00
81.00
74.00
70.00
76.00
66.00
70.00
63.00
67.00
70.00
76.00
71.00
76.00
76.00
72.20
Gains in per capita output
Constant
participation
Increasing
participation
If no
ageing
1.44
1.77
1.84
1.71
1.76
1.70
1.29
1.73
1.20
2.01
1.79
1.51
1.04
1.77
1.99
1.64
1.44
2.11
1.81
1.85
2.01
1.79
1.56
1.98
1.53
2.40
2.05
1.59
1.17
1.86
2.09
1.82
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
Cyprus
Czech Republic
Estonia
Hungary
Latria
Lithuania
Malta
Poland
Slovakia
Slovenia
Average
0.18
0.2
0.24
0.22
0.24
0.22
0.19
0.18
0.16
0.21
0.21
0.39
0.59
0.57
0.5
0.56
0.43
0.46
0.5
0.47
0.64
0.51
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.26
1.56
1.72
1.95
1.76
2.22
2.03
1.85
1.92
1.38
1.86
Japan
US
0.29
0.18
0.72
0.32
78.00
85.00
0.96
2.02
Note: NA: not available
Source: EEAG.
0.99
2.16
3.04
3.04
3.04
3.04
3.04
3.04
3.04
3.04
3.04
3.04
3.04
2.44
2.44