Consumption and the Permanent Income Hypothesis

Craig Burnside
Duke University
Fall 2009

1 A Simple Model of Household Consumption

Imagine a household with the instantaneous utility function u(Ct ) that wishes to maximize
its lifetime utility
X
T 1
t
U= u(Ct ); (1)
t=0

where 0 < < 1. Notice that the household has a …nite life, T periods in length.
The household gets endowment income in each period, Yt , in each period, begins its life
with some quantity of assets A0 , and can borrow or lend at the exogenous interest rate r in
any period. Thus it makes sense to model the household’s ‡ow budget constraint as

At+1 = At (1 + r) + Yt Ct ; (2)

for 0 t T 1. The household is constrained to not have debt at the end of its life: i.e.
AT 0.
You could imagine solving the household’s problem one of two di¤erent ways. One would
be to set up the Lagrangean

X
T 1 X
T 1
t
L= u(Ct ) + t [At (1 + r) + Yt Ct At+1 ] + AT (3)
t=0 t=0

where each period budget constraint is treated as a separate constraint, and has a separate
Lagrange multiplier, and we explicitly include the nonnegativity constraint on AT . Alterna-
tively one could iterate on (2) and obtain

A0 = (1 + r) 1 A1 + (1 + r) 1 (C0 Y0 )
= (1 + r) 2 A2 + (1 + r) 1 (C0 Y0 ) + (1 + r) 2 (C1 Y1 )
X
T 1
T
= = (1 + r) AT + (1 + r) (t+1) (Ct Yt ); (4)
t=0

and then write the Lagrangean

" #
X
T 1 X
T 1
t T (t+1)
L= u(Ct ) + A0 (1 + r) AT (1 + r) (Ct Yt ) + A T : (5)
t=0 t=0

1
 If we work with the …rst formulation we get the …rst-order conditions:
t 0
Ct : u (Ct ) = t, 0 t T 1 (6)
At+1 : t = (1 + r) t+1 , 0 t T 2 (7)
AT : T 1 = (8)

whereas for the second formulation we get the …rst-order conditions:

t 0 (t+1)
Ct : u (Ct ) = (1 + r) , 0 t T 1 (9)
T
AT : (1 + r) = (10)

Notice that (7) implies that

1 t
t = (1 + r) t 1 = = (1 + r) 0

for t T 1. Hence, we can rewrite (6) as

t 0 t
u (Ct ) = (1 + r) 0; 0 t T 1: (11)

The Lagrange multiplier on the non-negativity constraint for (3) is = T 1 = (1 +

(T 1)
r) 0, so (11) can be rewritten as
t 0
u (Ct ) = (1 + r)T 1 t
; 0 t T 1: (12)

On the other hand, notice that (9) and (10) can be combined to yield
t 0
u (Ct ) = (1 + r)T 1 t
, 0 t T 1: (13)

Since should equal (the marginal value of the constraint on AT should not depend on
how we setup the problem) we can see, from (12) and (13) that the …rst-order conditions are
identical.
Notice that regardless of which formulation we use, the model implies that the household
will want to maintain a constant marginal rate of substitution (MRS) between consumption
at two successive dates:
@U=@Ct+1 u0 (Ct+1 ) (1 + r) (t+2)
1
mt;t+1 = = 0
= (t+1)
= ; 0 t T 2: (14)
@U=@Ct u (Ct ) (1 + r) 1+r

1.1 An Interesting Special Case

An interesting case arises when = 1=(1 + r). We sometimes use the term rate of time
1
preference to describe = 1. Notice that when = 1=(1 + r) it follows that = r:
the rate of time preference is equal to the interest rate.

2
 Notice that in this special case (14) implies u0 (Ct+1 ) = u0 (Ct ) or Ct = C for some constant
C, for all t. Can we determine the constant C? Yes. Notice that if we use the fact that
Ct = C in (4), we have
X
T 1 X
T 1
(t+1) (t+1)
(1 + r) C = A0 + (1 + r) Yt : (15)
t=0 t=0

Notice that C is constant, and that the present value of consumption is equal to initial assets
plus the present value of income. This is the basis of the permanent income hypothesis in
that what is relevant for date t consumption is not date t income, but the household’s initial
assets and the present value of the income it expects to earn over its lifetime.
Of course, (15) can be simpli…ed to give
" #
r X
T 1
C= A0 + (1 + r) (t+1) Yt (16)
1 (1 + r) T t=0

where the right-hand-side is what we call permanent income. You can see that with this
de…nition, permanent income is essentially the annuity value of [ the household’s wealth plus
the present value of its current and future income ].
Romer consider the case where = 1 (i.e. = 0) and r = 0. In this case we cannot
directly use the formula given in (16) but we can work with (15) to solve for C = [A0 +
PT 1
t=0 Yt ]=T . The di¤erence between this and (16) is that income is not discounted, and,
because the interest rate is zero, the rate at which wealth is annuitized is just 1=T , where T
is the length of one’s life.
Notice what this means is:

1. Even if income is bouncing around over time, consumption does not track income–in
fact, in this special case, consumption is constant regardless of how volatile income is.

2. Savings moves around a lot if income is volatile, since it is just Yt Ct = Yt C. I.e.

savings is as volatile as income is.

1.2 More General Cases

Suppose 6= r. For example, suppose > r, so that the household tends to discount future
consumption more than a household with = r does. Notice that this means
u0 (Ct+1 ) 1+
0
= >1 =) Ct+1 < Ct :
u (Ct ) 1+r
The household will have a declining consumption pro…le. But notice that the path of con-
sumption will still be smooth since the MRS is constant for all t. In other words, consumption
will be declining, but along a very smooth path.

3
 In the opposite case where < r, the household, not surprisingly, has an increasing
consumption pro…le since < r implies u0 (Ct+1 ) < u0 (Ct ).

1.3 Empirical Implications and the Keynesian Model

In traditional Keynesian models of aggregate ‡uctuations, aggregate consumption is a simple
linear function of aggregate income: C = a + bY , with 0 < b < 1 being the marginal
propensity to consume out of income. This relationship followed from Keynes’belief that at
the individual level, the higher one’s income, the higher one’s saving rate would be.
When we consider data, one thing we can do is take a cross-sectional sample of households
and plot household consumption against household income. When this is done with US data
the diagram looks something like Figure 7.1a in Romer. This, indeed, looks consistent
with the Keynesian notion that the savings rate would increase with income, since when
C = a + bY , S=Y = 1 b a=Y .
If, instead of plotting household consumption against household income at one point in
time, we plotted aggregate consumption against aggregate income over time, the picture
would look something like Figure 7.1b in Romer, for US data. This suggests, in contrast to
the simple versions of the Keynesian theory in our textbooks, that there is a stable ratio of
C=Y over time. As you will see in the homework, explaining this requires thinking about
how growth over time a¤ects the analysis.
Finally, Romer’s Figure 7.1c shows a graph, like Figure 7.1a, where the data are divided
into two di¤erent socio-economic groups, say whites and blacks. Again, consumption appears
to obey the relationship C = a + bY , but a is smaller for the group I have labelled “black”
while b seems to be the same across groups.
We will now discuss whether these observations are consistent with our theory of con-
sumption being determined by permanent income.
Suppose you have some cross-sectional data on household consumption, Ci and household
income, Yi , for N households, so that i = 1, 2, : : : , N . We will de…ne transitory income, YiT ,
implicitly using the following identity:

Yi = YiP + YiT (17)

where YiP , permanent income, is de…ned by the right-hand side of (16). Our theory implies
that
Ci = YiP : (18)
We want to see if our theory is consistent with, say, Figure 7.1a, where household consump-
tion is plotted against household income, and a regression is used to …t a line through the
scatter plot of Ci on Yi .

4
 If we were to run a regression of Ci on Yi , to obtain Ci = a + bYi , we know, from simple
econometric theory that
^b = Cov(Ci ; Yi )=V ar(Yi )

^ = C ^bY :
a

Notice that we can use (17) and (18) to rewrite ^b as

P P T
^b = Cov(Ci ; Yi ) = Cov(Yi ; Yi + Yi )
V ar(Yi ) V ar(YiP + YiT )
V ar(YiP ) + Cov(YiP ; YiT )
= :
V ar(YiP ) + V ar(YiT ) + 2Cov(YiP ; YiT )
Romer argues that we should expect Cov(YiP ; YiT ) to be small. This makes a certain amount
of sense. Within our sample of household data we might expect some individuals to be having
a relatively low income year, so that their transitory income would be negative, and we might
expect some to be having a relatively high income year, so that their transitory income would
be positive. However, there’d be no particular reason to believe that high income individuals
would mostly be having a good year, while low income individuals were mostly having a bad
year, or vice versa.1 In this case we’d expect Cov(YiP ; YiT ) 0. In this case

^b = V ar(YiP )
V ar(YiP ) + V ar(YiT )
which implies that 0 < ^b < 1. Also, again using (17) and (18) we have

a
^ = C ^bY = Y P ^b(Y P + Y T )

= (1 ^b)Y P ^bY T

In a typical year, we would expect Y T 0, and even in a very good year, we would expect
T P
the scale of Y to be small relative to Y , hence it is reasonable to believe that a
^ > 0 in
most samples. So our theory is consistent with Figure 7.1a.
When we consider Figure 7.1c, the explanation for the slopes being similar for blacks and
whites is that the relative variance, V ar(YiP )=V ar(YiT ) is similar within the two groups. On
the other hand, consider a year in which Y T is approximately zero for both groups. Notice
^ (1 ^b)Y P . Since average income levels are higher for whites than for
that in this case, a
blacks, this is also true of permanent income levels, and this explains the higher intercept
for the white group.
To show that our theory is consistent with Figure 7.1b requires that we think about the
impact of income growth on the time series relationship between consumption and income.
We will examine this in the homework.
1
Most people might be having one or the other, say because we were looking at a boom or recession year.

5
2 A Model with Uncertainty
Now imagine that our household is uncertain about its future endowment income. In all
other respects we leave the model unchanged. At time 0, the household maximizes
X
T 1
t
E0 u(Ct );
t=0

subject to A0 given, AT 0, and the sequence of budget constraints

At+1 = At (1 + r) + Yt Ct ; 0 t T 1:

When we optimize under certainty, we can formulate the problem as a Lagrangean as before,
or use dynamic programming. The important thing to keep in mind is that the household
cannot choose the entire path of consumption at time 0 because of uncertainty. It can
only choose C0 and contingency plans for Ct , t 1, where the plans are contingent of the
realizations of Yt .
Since we haven’t changed anything in the model, other than to make the income stream
uncertain, we end up with the same …rst-order conditions, but any intertemporal conditions
hold in expectation as opposed to holding with certainty. So we end up with the following
conditions
t 0
t = u (Ct ) (19)
t = (1 + r)Et t+1 (20)

These equations imply that

u0 (Ct ) = (1 + r)Et u0 (Ct+1 ):

Notice that this means the MRS is not constant, however, the one-step ahead forecast of the
MRS is constant:
Et u0 (Ct+1 ) u0 (Ct )=[ (1 + r)] 1
Et mt;t+1 = 0
= 0
= :
u (Ct ) u (Ct ) 1+r
Deviations of the MRS from its conditional mean, 1=(1 + r), are, therefore, white noise. I.e.
we have
1
mt;t+1 = (1 + r) + t+1 :

2.1 A Special Case, Again

Notice that if r = , we have u0 (Ct ) = Et u0 (Ct+1 ), so that the marginal utility of consumption
is a random walk (or martingale). I.e. it is a time series process, xt , such that Et xt+1 = xt .
In other words the one-step-ahead forecast of xt is just xt .

6
2.2 The Random Walk Hypothesis
Suppose the utility function is quadratic: i.e. u(C) = C (a=2)C 2 , implying that u0 (C) =
1 aC. As a result, if we consider the case where r = , we have 1 aCt = 1 aEt Ct+1 ,
implying that
Et Ct+1 = Ct :
I.e. consumption is a random walk. Thus we can write Ct = Ct 1 + t . This result is not
too surprising. In the deterministic case we had the result that Ct = C for all t. Here,
the household chooses to have a consumption path which has no predictable changes. When
uncertainty is resolved, the household will adjust its consumption, but it always does it in a
way that implies that any future changes are unpredictable.
In a more general setting, of course, we have

u0 (Ct ) = (1 + r)Et u0 (Ct+1 ): (21)

Does a version of the random-walk hypothesis hold in this case? It does, as long as we
are willing to work with …rst-order approximations to the …rst-order conditions. Notice, for
example, that we could approximate the right-hand side of (21) in the neighborhood of Ct .
In particular we could write

u0 (Ct ) (1 + r)u0 (Ct ) + (1 + r)u00 (Ct )Et (Ct+1 Ct ):

This would imply that

r u0 (Ct )
Et Ct+1 Ct :
1+r u00 (Ct )
This means that if = r, the random walk hypothesis will approximately hold for more
general utility functions.

3 Testing the Permanent Income Hypothesis

3.1 Hall’s Test
Early tests of the PIH used an in…nite horizon version of our model with uncertainty. To
extend the problem to the in…nite horizon we must impose a condition that is the analog of
AT 0. This condition is
lim (1 + r) t At = 0
t!1

which is discussed further in the homework. Our household’s problem will be to maximize
X
1
t
E0 u(Ct );
t=0

7
subject to A0 given, limt!1 (1 + r) t At = 0, and the sequence of budget constraints

At+1 = At (1 + r) + Yt Ct ; t 0: (22)

We end up with the same …rst-order conditions as above, (19) and (20), and therefore, it
continues to be the case that (21) holds. Furthermore, if we assume quadratic utility and
= r we get
Ct = Et Ct+1 : (23)
Notice that (23) means that consumption changes between t and t + 1 are unforecastable
at time t. Hall’s tests of the model essentially involved seeing whether that was true. He
measured Ct+1 Ct and regressed it on variables dated t and earlier. According to our
theory, none of these variables should have power in predicting Ct+1 Ct and should show
up as insigni…cant in the regressions. Hall found that this was true for a large number of
variables, but he also found that some variables did help to predict consumption changes,
notably stock market data. This led him to reject the model.

3.2 Flavin’s Test

When we solved the model with perfect foresight we had the condition Ct = C for all t. We
combined this condition with the lifetime budget constraint to get an explicit solution for
C in terms of A0 and the present value of the future income stream. We will do the same
thing here. Notice that if we iterate on the budget constraint, (22), starting at time t, and
we impose limt!1 (1 + r) t At = 0, we get
X
1 X
1
(j+1) (j+1)
(1 + r) Ct+j = At + (1 + r) Yt+j : (24)
j=0 j=0

If we take the conditional expectation of both sides at time t we have

X
1 X
1
(j+1) (j+1)
(1 + r) Et Ct+j = At + (1 + r) Et Yt+j : (25)
j=0 j=0
P1 (j+1)
Using (23), and the fact that j=0 (1 + r) = 1=r, we can write this as

X
1
(j+1)
Ct = r[At + (1 + r) Et Yt+j ]: (26)
j=0

This equation allows us to solve for the change in consumption over time, Ct Ct 1 . Notice
that if I substitute the fact that

At = At 1 (1 + r) + Yt 1 Ct 1

8
into (26) I get
X
1
(j+1)
Ct = r(1 + r)At 1 + rYt 1 rCt 1 +r (1 + r) Et Yt+j : (27)
j=0

I can also get an expression for (1 + r)Ct 1 using (26) dated back one period, and multiplied
through by (1 + r):
X
1
(j+1)
(1 + r)Ct 1 = r(1 + r)At 1 + r(1 + r) (1 + r) Et 1 Yt 1+j : (28)
j=0

Subtracting both sides of (28) from the equivalent sides of (27) we have
X
1
(j+1)
Ct (1 + r)Ct 1 = r(1 + r)At 1 + rYt 1 rCt 1 +r (1 + r) Et Yt+j
j=0
" #
X
1
(j+1)
r(1 + r)At 1 + r(1 + r) (1 + r) Et 1 Yt 1+j :
j=0

Notice that the rCt 1 on the LHS cancels with rCt 1 on the RHS, and that the two
r(1 + r)At 1 terms cancel with one another, so that we have
X
1 X
1
(j+1) (j+1)
Ct Ct 1 = rYt 1 +r (1 + r) Et Yt+j r(1 + r) (1 + r) Et 1 Yt 1+j :
j=0 j=0

Notice, also that rYt 1 cancels with the …rst term in the second in…nite sum, which is
rEt 1 Yt 1 . With some careful rewriting, we can then combine the two sums and write
X
1
(j+1)
Ct Ct 1 =r (1 + r) (Et Yt+j Et 1 Yt+j ): (29)
j=0

Since we know that the change in consumption is white noise, i.e. Ct Ct 1 = vt , where
vt is some white noise process, we now have an explicit expression for vt . It is the change,
between dates t 1 and t, in the expectation of lifetime income from date t forward.
Flavin’s tests are based on estimating a joint model of consumption and income that
nests the PIH. The null hypothesis that the PIH is true can then be tested. She assumes
that
(L)Yt = 0 + t

2 p
where (L) = 1 1L 2L pL . In other words, she assumes that Yt is a p-th
order autoregressive process. She then writes

Yt = (L) 1 ( 0 + t)
= ~ 0 + (L) t
= ~0 + 0 t + 1 t 1 + 2 t 2 + ;

9
where ~ 0 = 0= (1), and (L) = (L) 1
= 0 + 1L + 2L
2
+ . Notice that means

Yt Et 1 Yt = 0 t

Et Yt+1 Et 1 Yt+1 = 1 t

and, by extension,
Et Yt+j Et 1 Yt+j = j t:

Hence, from (29) we have

" #
X
1
(j+1)
Ct Ct 1 = r (1 + r) j t: (30)
j=0

P1 (j+1)
To simplify notation we will de…ne =r j=0 (1 + r) j and write Ct Ct 1 = t.

Flavin’s test of the PIH essentially involves writing down the following model of con-
sumption and income

Ct = t + 0 + 1 Yt + 2 Yt 1 + ut (31)
Yt = 0 + 1 Yt 1 + 2 Yt 2 + t: (32)

In the equation for output, (32), the order of the autoregression, p, has been set equal to
2. In the consumption equation, (31), several terms have been added relative to what the
theory predicts, which is Ct = t . The error term ut is added to allow for the fact that
consumption may not be measured perfectly in the data, so that even if the PIH is correct,
observed Ct may di¤er from t. The term 0 is added to allow for the possibility that
consumption grows over time. Under the null hypothesis that the PIH holds, the coe¢ cients
on the Yt and Yt 1 terms should be zero; i.e. we can test the PIH by testing whether
1 = 2 = 0.
To estimate the model one has to use instrumental variables, since the error term in the
…rst equation, t + ut , is correlated with Yt . Of course, lagged changes in income would
be natural candidate instruments because as long as Yt is predictable, lagged values of Yt
should be correlated with it (therefore they will be relevant instruments) and, also, lagged
values of Yt should be uncorrelated with t (therefore they will be valid instruments).
If 1 6= 0 or 2 6= 0, as Flavin found, we usually describe this as a situation where
consumption is excessively sensitive to changes in income. But notice one important thing.
Consumption, in this case, is excessively sensitive to changes in income that were predictable
at time t 1, since the unpredictable components of income are included in the error term
in the regression.

10
3.3 Campbell and Mankiw’s Test
Campbell and Mankiw proposed a di¤erent test of the PIH, where they had in mind a speci…c
alternative model to the PIH. In particular they assumed that some set of consumers that
represent a fraction 1 of aggregate consumption, do behave according to the PIH. So for
these consumers Ct Ct 1 = vt , a white noise process.
On the other hand, they supposed that there is another group of consumers, representing
a fraction of aggregate consumption, that does not behave according to the PIH. This
group of consumers simply absorbs any change in its income through an equal change in
its consumption: Ct Ct 1 = Yt Yt 1 . Notice that this means that for this group of
consumers St St 1 = 0 (i.e. there is no change in this group’s savings). One motivation
for this hypothesis is that some consumers are liquidity constrained, i.e. they have hit
some constraint on their borrowing, that does not allow them to dissave in order to smooth
consumption. This is not, of course, the only motivation, but it is a standard one.
In this case, we have, at the aggregate level:

Ct Ct 1 = (Yt Yt 1 ) + (1 )vt :

Again, we cannot directly estimate this equation by least squares, since we would expect
Yt Yt 1 to be correlated with vt . However, we could use instrumental variables as we
described for the Flavin test. Campbell and Mankiw found that using lagged Yt as an
instrument does not work very well because it does not seem to have much correlation with
Yt (i.e. it is a valid but not very relevant instrument). They used lagged consumption
changes instead, and found that was positive and signi…cantly di¤erent than zero. You
can see that their test is very similar in spirit to Flavin’s but it provides a direct interpretation
of the coe¢ cient on Yt as the fraction of consumers who are not behaving according to the
PIH.
There is an interesting discussion in Romer of Shea’s work. He studies household data in
order to try to …nd a good interpretation of what ^ > 0 means. In particular, he tries to see
whether it makes sense to interpret ^ > 0 as the fraction of household’s that are liquidity
constrained. You should read this part of Romer.

11
4 Consumption’s Smoothness
Modeling the Income Process When we described Flavin’s test we assumed that income
was stationary. Of course, this is likely to be counterfactual since real income tends to rise
over time for most individuals and in the aggregate. However, the assumption of stationarity
can be generalized to that of trend-stationarity where
Yt = (t) + (L) t (33)
and is a deterministic function of time. Notice that this does not change the fact that
Et Yt+j Et 1 Yt+j = j t . Hence, allowing for deterministic trends in income would make no
di¤erence as far as the model’s predictions about Ct are concerned.
Deaton (on p. 105) argues in favor of a di¤erence-stationary representation of income,
arguing that when future income is forecasted, intuitively one should expect the forecast
error variance to rise with the forecast horizon. Trend-stationary models are not consistent
with this, whereas di¤erence-stationary models are. So he argues for the representation
X1
Yt = vt = j t j: (34)
j=0

Here, by having > 0 we can allow for upward drift in income levels, since is the mean of
the change in income over time. The polynomial (L) is assumed to be consistent with vt
being a stationary process.
To understand Deaton’s point consider a simple example of a trend-stationary process:
a …rst-order autoregressive model–an AR(1) model— in which (1 L)(Yt a bt) = t ,
2
with j j < 1. Notice that this means Yt = a + bt + t + t 1 + t 2 + , so that
2
(L) = 1 + L + L2 + . We can construct a simple example of a di¤erence-stationary
process by letting = 1 in my AR(1) example: i.e. we can let (1 L)(Yt a bt) = t.

Notice that this means Yt b = t . So in this example, (L) = 1.

In our trend-stationary example if I forecast output j periods ahead my forecast error is
f 2 j 1
t;t+j = Yt+j Et Yt+j = t+j + t+j 1 + t+j 2 + + t+1 .

Notice that the variance of this forecast error is

2j
f 2 4 2(j 1) 2 1 2
Var( t;t+j ) = [1 + + + + ] = 2
;
1
2 2
which converges to a constant =(1 ) as j ! 1. On the other hand, for the di¤erence-
stationary example if I forecast output j periods ahead my forecast error is
f
t;t+j = Yt+j Et Yt+j = t+j + t+j 1 + t+j 2 + + t+1 .
f 2
Notice that the variance of this forecast error is Var( t;t+j ) = j which limits to 1 as
j ! 1.

12
Why the Model of Income Matters The model (34) implies that Et Yt+j Et 1 Yt+j is
given by

(Et Et 1 )Yt+j = (Et Et 1 )(Yt 1 + Yt + Yt+1 + + Yt+j )

= ( 0 + 1 + + j ) t:

Hence,
X
1
(j+1)
Ct = r (1 + r) (Et Et 1 )Yt+j
j=0
X
1
(j+1)
= r (1 + r) ( 0 + + j) t
j=0
or
"1 #
r X X
1 X
1
Ct = (1 + r) j 0 + (1 + r) j
1 + (1 + r) j
2 + t
1 + r j=0 j=1 j=2
"1 #
X
= (1 + r) j j t : (35)
j=0

We can de…ne a polynomial b(L) implicitly using b(L)(1 L) = (L). Notice that
2 2 2
b0 + b1 L + b2 L + b0 L b1 L = 0 + 1L + 2L + . Hence, 0 = b0 , but
j = bj bj 1 for j
1. So we can write (35) as
" # " #
X1
r X
1
Ct = b0 + (1 + r) j (bj bj 1 ) t = (1 + r) j bj t: (36)
j=1
1+r j=0

This result is equivalent to (30), but should not be surprising since, with an abuse of notation,
(L) = (1 L) 1 (L) is the moving average polynomial appropriate for the level of Yt .
Deaton illustrates an interesting puzzle relating to the choice of how to pick the time
series representation of Yt . He uses aggregate labor income data to estimate autoregressive
special cases of (33) and (34):

(L)[Yt (t)] = t (37)

(L)( Yt ) = t: (38)

In estimating (37) he uses an AR(2) representation and …nds (L) = 1 1:42L + 0:45L2 . In
estimating (38) he uses an AR(1) representation for output’s growth rate and …nds (L) =
1 0:44L. Notice that these two estimated models imply quite similar looking representations
for the level of Yt since the estimated versions of the two models imply:

(1 1:42L + 0:45L2 )Yt = deterministic part + t

2
(1 0:44L)(1 L)Yt = (1 1:44L + 0:44L )Yt = 0:56 + t :

13
 Using a real interest rate of r = 0:01, Deaton shows that the trend-stationary model
delivers the solution Ct = 0:28 t while the di¤erence-stationary solution delivers the solu-
tion Ct = 1:77 t . Why are the models so di¤erent, despite having similar autoregressive
representations for the level of Yt ?
One way to see why is to ignore the deterministic trend and consider a second-order
autoregressive representation for Yt , (L)Yt = t , factored as (1 L)(1 L)Yt = t . Notice
that with this representation, Yt = (L) t = (1 L) 1 (1 L) 1 t , with

2 2
(L) = (1 + L + L2 + )(1 + L + L2 + )
2 2
= 1 + ( + )L + ( + + )L2 +

so that
j
X j+1 j+1
j i i
j = = :
i=0

Using the solution (30) this implies

r(1 + r)
Ct = t = t:
(1 + r )(1 + r )

Notice that
1+r
lim =
!1 1+r
so that when there is a unit root, the coe¢ cient on will be greater than 1 for > 0.
So if you impose = 1 (as in the di¤erence stationary model) and …nd > 0 (i.e. there is
positive serial correlation in the growth rate of output), you will necessarily …nd that >1
(as Deaton does in his example). But this means that in period t, Ct should rise by more
than t , which is the innovation in the level of income. This would mean that consumption—
which is the same as permanent income— should be more conditionally volatile than income:
2 2 2
Vart 1 ( Ct ) = > Vart 1 ( Yt ) = . Subject to a restriction on , it is also possible to
…nd Var( Ct ) > Var( Yt ) in the case where = 1.
Since— in U.S. data— growth rates of income are positively serially correlated one will
tend to …nd > 1. On the other hand, one also …nds changes in consumption that seem
to less volatile than changes in income. This suggests that consumption in the data is
excessively smooth.
Why does the trend-stationary model have such di¤erent implications for ? Although
= 1 implies > 1, it turns out that does not need to be much less than 1 for to
be quite small. Notice that our di¤erence stationary model implies that = 0:44, = 1
and r = 0:01, so = 1:772. On the other hand, the trend-stationary model, with the AR

14
polynomial 1 1:42L + 0:45L2 can be factored as (1 0:48L)(1 0:94L). Thus, = 0:48,
= 0:94 implying = 0:272.
More generally, notice that
1+r
= 1 ) :
1+r
r
= 1 ) =1
1+r

When is signi…cantly bigger than 0, but signi…cantly smaller than 1, (as in our case) and
r is small, (1 + r)=(1 + r ) will be much bigger than 1, whereas 1 r=(1 + r ) will be
very close to 1.

15