Universitat Pompeu Fabra Introduction to Macroeconomics 2022-23

Prof: Teresa Garcia Milà, María Gundín, Alberto Martín.

Problem set 1

SECCTION A.
The following statements are false. Explain why.
a/ In an economy you cannot simultaneously observe an increase in the price level and a decrease
in the inflation rate (both calculated using the CPI – Consumer Price Index). Propose a numerical
example to illustrate your answer.
False: prices may increase but at a lower rate than the previous period → higher price level +
lower inflation rate. Example:
Year 2001 2002 2003
Price 20 30 35

The prices increase between 2001 and 2003 but π =50% and π =16.6%
02 03

b/ If the Unemployment rate in an economy is 20% the employment rate is 80%.

Unemployment rate is defined as the share of unemployed in the labor force (u = U / L)
Employment rate is defined as the share of employed in the working-age population (e = E / WAP)

In addition:
The unemployment rate only takes in account people actively looking for a job. In Europe, this
normally means that jobseekers are registered at a government's office in order to claim
unemployment benefits. Countries with little or no unemployment benefits typically have
underestimated unemployment figures.
A lot of people are not employed but are not "unemployed" because they're not looking for a
job. Being infants, home stay parents, retired, students, etc. Also "discouraged workers" who
were looking for work but have dropped out of the workforce because they've found it too
difficult to find a job.

c/ If nominal economic growth in a mature and developed economy is 10%, we can affirm this is
good news since it implies a generous increase in production and it is expected a net creation of
jobs.
Nominal GDP is the sum of the quantities of final goods produced multiplied by their current
prices. This definition just makes clear that nominal GDP can increase over time for two reasons:
● The production of most goods increases over time.
1
● The prices of most goods also increase over time.
If our goal is to measure production and its change over time, we need to eliminate the effect of
possible variation of prices on our measure of GDP. That’s why real GDP is constructed as the sum
of the production of final goods multiplied by constant (rather than current) prices.
Thus, nominal GDP could increase due to an increase in prices while real GDP could remain
constant or decline if production is not increasing at the same time.

d/ Deflation means a decrease in the inflation rate.

Deflation is a decrease in the general price level of goods and services. Deflation occurs when
the inflation rate falls below 0% (a negative inflation rate). This allows one to buy more goods and
services than before with the same amount of currency. Deflation is distinct from disinflation, a
slow-down, a decrease in the inflation rate, i.e. when inflation declines to a lower rate but is still
positive.

SECTION B

1B Assume that nominal GDP of an economy is €200 billion in 2012, and €220 billion in 2013. If
during this period the inflation rate was 1%, what was the growth rate of real GDP?
Nominal GDP growth rate= (220 – 200) / 200 = 0,10 = 10%
Inflation rate = 1%
Real GDP growth rate = Nominal GDP growth rate – Inflation rate = 10% – 1% = 9%

2B. Consider a country with the following characteristics: There are 35 million individuals, 25
million of working age, of which 14 million are employed, 6 million are searching for a job, 1,5
million have stopped searching for a job within the last two months, and the remaining do not
want to work.

a/ Calculate the unemployment, the participation and the employment rates.

Working-age population (WAP) = 25 M.
Employed: E = 14 M. Unemployed: U = 6 M. Labor force: L = E + U = 20M
Unemployment rate: u = U / L = (6 / 20 ) * 100= 30%
Participation rate: PR = L / WAP = (20 / 25)* 100 = 80%
Employment rate: e = E / WAP = (14/ 25) * 100 = 56%

2
b/ Assume now that 200, 000 new jobs have been created, none of the previous jobs have
disappeared, and 700,000, out of the 1.5 million that have stopped looking for a job, join the labor
market. Taking this fact into account, obtain both the unemployment and employment rates, and
comment on your results compared with the previous situation. Why is it so important to consider
the data for Employment when analyzing the evolution of the unemployment rate?
Employed: E = 14,2 M. Unemployed: U = 6,5 M. Labor force (active population): L = E + U =
20,7M
Unemployment rate: u = U / L = (6,5 / 20,7)*100 = 31,4%
Employment rate: e = E / WAP = (14,2 / 25)*100 = 56.8%
Unemployment rate increases, but also does employment rate. So, the explanation lies on the
increase in participation rate. New jobs were not enough for employing people joining the
market.

3B. Look at the data, from Eurostat, in graph below. Explain the behavior and relationship of the
three variables shown. Split your analysis in two periods: [2005- 2007) and [2007 – 2013)

[2005- 2007)
• There is economic growth since we can see a positive variation of GDP
• Employment increases and unemployment decreases: there has been a net creation of jobs, and
no change in the active population.

[2007 – 2013)
• At the beginning of the period economic growth decreases, it even turns negative (economic crisis)
twice. By the end of 2012 crisis is still here.
• Employment decreases and unemployment increases, there was net job destruction: less people
work and, if active population did not vary, this increases unemployment

3
 3C. Looking at the graph below, say if the following statements are correct or not, if they are not, explain
why

Per capita GDP relative to the EU average

Source: European Comission

● Along the period shown Italians are the wealthier inhabitants TRUE
● Italy was having an economic recession most of the years shown, in general GDP per capita was
decreasing. FALSE, we cannot say, it can be the case that it was growing but less than other EU
countries.
● The GDP per capita of the three poorest countries was growing above the EU average growth rate
TRUE

SECTION C. Readings

1. Go to the reading GDP and capturing the benefits of the Internet economy by Charles R. Hulten,
Leonard Nakamura 02 June 2017 Published on VOX, CEPR’s Policy Portal (http://voxeu.org)

A. According to the article internet and software applications improved the utility of consumers,
who now have more services disposable for consumption. In the economies where those
software applications were generated there have been a lot of new services produced.
However, this large number of new services produced does not seem to be considered in GDP.
According to this information, is economic activity well measured by GDP? Explain.
B. …” increases in effective information and product quality allow the consumer to get more utility
from a given amount of expenditure...” In which way does this improve welfare even without
implying an increase in GDP?
The GDP as a measure of growth of an Economy only measures income and production, and it does so at
market prices. People might not only be interested in the growth of income and production, but rather in
whether agents are better off from one year to the next – consumer surplus as a potential variable that
could indicate such improvements. However, GDP does not measure such improvements if they cannot
be calculated and measured at market prices. The Digital Economy provides a good example. Many
4
 agents do experience improvements in their life because they can e.g. access data, newspapers, and so
on for free. As long as agents do not have to pay for such services (although most likely quite a few of
them would) this improvement does not show up in GDP changes.

2. Read the paragraphs in blue from the reading “The multiplier of the United States’ largest scholarship
programme “, April 2022.

2.a. In the reading the authors say that these grants will cause an increase in the GDP, both in the
short-run and in the long-run. Can you think about the reasons of this long-run (more than ten years
ahead) economic growth for the US?

We can think about education as an increase in the habilities of children who will become more
efficient workers in the future. So in that future, one hour of work of a qualified worker produces a
higher value (higher GDP), that one hour worked by a less qualified person. The first one gets a higher
wage than the second one (so higher GDP)

Also, higher GDP, can pay higher amount of taxes with no need to raise taxes, so more public money
that can be used to stimulate growth.

2.b Regarding the short-run economic growth, the author talks about: “one-for-one increase in
personal income”; and also about giving money to low income students, who have a high marginal
propensity to consume. Using our model describing the goods market, and focusing on the
consumption multiplier, do these grants have a larger effect on GDP than if they were distributed
among wealthier students? Why?

A grant means an increase in disposable income for students, increase in personal transfers (like a decrease
in taxes)

Lower income students can save a very small proportion of any increase in their income, there is nearly no
room for saving. This means that they use a large proportion, for consumption matters (large c1). In the
case of wealthier students, their capacity of saving is larger facing an increase in their income. So, their
marginal consumption, c1 is lower, the spend in consumption a smaller proportion of that extra income.

!
Assuming, for example, the very simple multiplier !"#!

!
Y= !"#!
* (autonomous spending).

An increase in Autonomous spending is “more multiplied” if C1 is large, so it creates a higher increase in

Y, income, GDP.

5
SECTION D: PROBLEM
1. Consider an economy described by the following equations:
(consumption) C = 245 + c1 Yd (investment) I = 180 mill €
(taxes) T = 200 mill € (government expenditures) G = 200 mill €
1.1 Given that the marginal propensity to consume is equal to 0.75, find the equilibrium values for GDP,
disposable income, consumption, saving (private and public saving). Compare total saving with
investment. Find the multiplier for this economy.
Y = C+I+G = 245+0.75 Yd +I+G = C+I+G = 245+0.75 (Y-T ) +I+G
Y = 1900 Yd = Y-T = 1700 C= 245 + 0,75 *1700 = 1520
Private savings = Y-T-C = 180 , Public saving = T-G = 0 , Total saving = 180 + 0 = 180 = Investment
Multiplier = 1/(1-0.75) = 4

1.2 Assume that the marginal propensity to consume goes down to 0.6. Answer the same questions as in
1.1.
Y = 1262,5 Yd = 1062,5 C = 882.5
Private savings= 180 Public saving = 0 Total saving = 180 + 0 = 180 = Investment Multiplier = 2,5

1.3 If the marginal propensity consume goes back to 0.75 and government expenditure increases up to
1.3 mill € without changing taxes, find the new GDP of equilibrium for the economy. Calculate saving
(private and public)
Y = 245 + 0.75 (Y-200) + 180 + 201.3 → Y = 1905.2
o ∆Y = 4 ∆G → ∆Y = 4*1,3 = 5,2
1.4 If the fiscal expansion of 1.3 million € is financed with a tax increase of 1.3 million €, is the GDP of
equilibrium any different from the one in section 1.3?
Y = 245 +0.75(Y – 201.3) +180 +201.3
Y=1901.3 which is lower than in the previous point

1.5 Explain the different effect of this increase in public expenditure policy (1.4) along with an increase in
taxes, with the expansionary fiscal policy in 1.3, with the exact same increase in G, financed with
government debt.
Increase in taxes decreases disposable income (proportional to marginal propensity to consume) =>
decreases consumption. Then the increase in GDP due to fiscal expansion for case 1.4 is lower than in 1.3.
1.6 Compare 1.4 with the initial situation in 1.1 and explain.
Y = 245 + 0.75 (Y-201.3) + 180 + 201.3 → Y = 1901.3
o ∆Y = 4 [∆G – c1 ∆T] → ∆Y = 4 [1,3 – 0,75*1,3] = 4 [1,3*0,25] = 1,3
That is, if ΔT = ΔG, ΔY = ΔG: the fiscal multiplier with balanced budget is equal to 1; the tax increase
cancels the multiplier effect on consumption, the increase in the demand for G remains. The negative
effect of ΔT is less than the positive effect of ΔG because c1 that multiplies ΔT is less than 1: one unit
less of disposable income reduces consumption only in c1 units.
6
Universitat Pompeu Fabra Introduction to Macroeconomics 2022-2023

Problem set 2

Section A. Problems

1. In 2020, the coronavirus and its consequences such as the lockdown translated into a decline in the
consumption of families. How could the lockdown influence autonomous consumption and the
propensity to consume? Represent graphically the consumption function in 2019 and that by the end
of 2020. Explain briefly the implications on the short term GDP, income, of that economy. How will
that change influence the multiplier?
Z Z
Y Y 45degrees line
a
<
a
n
d

Y Y
With Covid two changes in the consumption function can be identified. The first is a decrease in c0 as
confidence of consumers decreases and uncertainty about the future increases, triggering consumers
to decrease their consumption of durables. The second is a decrease in the marginal propensity to
consume, c1 because for the same income, some services are not available (restaurants, traveling for
vacation) and thus cannot be consumed. The result is a higher marginal propensity to save.
The red line represents the aggregate demand. The first graph corresponds to 2019. The graph on
the right, representing aggregate demand after Covid, has a lower autonomous spending, because
of the decrease in c0 and a lower slope, because of the decrease in the marginal propensity to
consume c1.

2. Focusing just on the GOODS Market, so far we have assumed that the fiscal policy variables G (public
expenditure) and T (taxes) are independent of the level of income. In the real world, however, this is
not the case. Taxes typically depend on the level of income and so tend to be higher when income is
higher.
Let us take as reference Section D: Problem from Problem Set 1 of last week. We want to think now
about the situation of a more realistic economy in terms of taxes.

Instead of taxes being constant (exogenous T), assume now that they depend on income such that tax
collection T = t*Y, being t the tax rate (%). In that economy, exogenous taxes, T, were 200€, and
income was 1900€.

2.a. Under those circumstances, which tax rate (t) would the economy need for tax revenues to be
maintained at 200€?

t = 10.5%
2.b. Go now to the information in 1.3 of the problem in PS1 and assume that taxes are endogenous (using
the tax rate “t” you have just obtained) and calculate the new GDP, after that increase in public spending.
Compare this new GDP with the one under exogenous taxes (1905.2€), explain.

Y = 245 + 0.75 (Y – 0,105Y) + 180 + 201.3

Y= 626,3 + 0,75 (0,895Y) à Y - 0,67125 Y = 626,3 0,32875Y = 626,3 Y = 1905,09 mill€
(versus 1905,2 mill €)

As expected, the new GDP after the increase in public spending is lower now (0,1 million € less) under
endogenous taxes. That increase in G, increased income and a part of that higher income goes to pay taxes,
the highest the income the highest taxes collected (see larger explanation below)

2.c Calculate the multiplier with endogenous taxes, and compare it to the multiplier obtained last week
with exogenous T (m=4). Explain

m=1/(1-c(1-t))= 3.04 < 4

Having endogenous taxes in the model implies a lower income multiplier. T increases proportionally to the
increase in Income. Hence, the multiplier is lower given that part of the income is absorbed by the taxes.

2.d. In a general form, for the simple economy with exogenous taxes, investment and Public spending G, the
!
multiplier is
!"#!

Show the general form of this multiplier for the case of endogenous taxes T = tY

Y = C + I + G = c0 + c1 (1 - t) Y + I + G .
Solving for Y,
Y = 1 / [ 1 - c1 (1 - t) * [ c0 + I + G ] .

!
Y= (𝐶𝑜 + 𝐼 + 𝐺) (!" #! - #! .)
The multiplier!!!

The multiplier tells you how much output changes when there is an exogenous change in spending (i.e.
change in c0 , I, or G). The multiplier effect arises because increases in spending translate to increases in
income for some households, who then want to increase their spending, which leads to further increases
in income, etc.
This means that when there is an increase of 1€ in the autonomous spending, the output will increase by
!
(!" #! - #! .)

This consumption increases less for a given change in autonomous spending, and so does aggregate
demand. Part of the increase in income is collected in taxes that do not contribute to an increase in the
aggregate demand under the assumption that G is constant.

When taxes depend on income, the economy is less responsive to changes in autonomous spending. As
said in 2.3 the multiplier is smaller here.

This is because some of the change in income will be paid as tax, so production will not increase by the
same amount as income (Y) and production will increase less when taxes depend on income than when
 they do not. The economy responds by less to changes in autonomous spending. The reason is that taxes
adjust to offset some of the effects of any income change. For example, if income Y increases, taxes T =
t1Y increase as well, so that disposable income Y - T increases by less than it would in the standard model.
Thus, consumers increase their consumption by less.

ANOTHER EXPLANATION:

This fiscal policy acts as an automatic stabilizer because it counterbalances (equilibra, suaviza)
movements in GDP: when GDP grows, tax collection increases, disposable income does not grow in the
same proportion as GDP (part of that change in GDP is paid in taxes) --> aggregate demand increases in
proportion less than GDP and so does production and the equilibrium output/income.

growth slows down during economic crisis --> tax collection decreases --> it softens GDP decline.

3. In late 2017, we could read in The Guardian: …”Responding to a survey commissioned by Positive
Money just before the June election, 85% were unaware that new money was created every time a
commercial bank extended a loan, while 70% thought that only the government had the power to
create new money….”

3.1 Explain briefly: …”new money was created every time a commercial bank extended a loan…”
As studied in class, once a commercial bank has enough money to meet its reserve requirements, it can
lend any remaining amounts to individuals and corporations in need of “credit”. Banks, however, do
not simply allow for borrowers to obtain the entire amount of the loan and walk away with it. Instead,
once approved, the loan amounts are (often in periodic installments) deposited into a borrower’s
checking account, which increases the M1 money supply (M1 = banknotes, coins, checks, checking
accounts etc., which are all very “liquid”, i.e. easily convertible into cash and usable as a “medium of
exchange” for the purchase of goods/services).
3.2 What can you say about: …” only the government had the power to create new money….”. IS THIS
CORRECT?
It seems that 70% of those answering the survey understand that the decisions about money, money
supplied, interest rates… are in the hands of the ministry of economy, or any other government
agency. This is not correct, since it is the Central Bank (usually independent from the aforementioned
government), the authority deciding on the amount of money supplied, created…)
In other words:
At first glance, the statement is quite ambiguous (usage of the word “government” is quite imprecise),
illustrating, in a sense, public misconceptions regarding the governance structure of central banks (the
main regulators of money/financial markets, deciding about the money supply).
Central banks (not the Government) control a portion of the money supply (through an array of
instruments, such as the policy rate/open market operations/reserve requirements), they remain
independent from finance ministries and other government agencies (which serves to ensure their
overall credibility).
At the same time, the statement alludes to the fact that central banks do not have sole/perfect control
of the money supply, because (for example) they do not directly control: i) household decisions re:
their deposits to/withdrawals from financial institutions and ii) private lenders’ decisions of whether or
not to lend money to private borrowers.
 4. The graph below shows the evolution of the reserve coefficient for China along time.

4.1 Explain the expected impact of that reserve coefficient along time on the Overall Money Supply
In the graph we observe the reserves requirement coefficient. A monetary tool in the hands of the Central
Bank.

Reduction in reserve requirement increases the money multiplier and therefore the amount of money that
can be loaned out by commercial banks for a given value of the monetary base (or high-power money). This
policy is aimed at banks that lend mainly to businesses and domestic economies. This measure makes it
easier for them to borrow money and invest more, or increase consumption, which may increase the
demand and the GDP in the short run.
4.2 Knowing that in the European Central Bank case, this coefficient has always been between 2% and 1%,
if we compare the money multiplier in both economies, what can be said?
In the European case the reserve coefficient is much smaller (1%) and therefore the money multiplier is, in
principle, much larger. Therefore, it is true that the ECB has a potentially larger multiplier

Let us see a numerical example, considering that in both economies the public holds a fraction c = 0.6 of
cash. We consider also that the commercial banks in both countries want to keep the minimum reserves
possible, let us check the value of multipliers in January 2022
1
• China θ= 0,115 mm=
0,6+0,115(1−0,6)
= 1,5479
1
• Europe: θ= 0,1 mm= = 1,6556
0,6+0,01(1−0,6)

Considering Ms = mm* Hs. --> if there is an increase in Central bank money of 1000 € / yuans relatively

o In China the overall Ms will increase in 1.547,9 yuans

o In the Eurozone the overall Ms will increase in 1.655, 6 €

4.3 Is there any other way for the ECB to increase the money Supply?
o Decrease the reserve´s coefficient (although being 1% nowadays the margin for a decrease in small)
o open market operations: CB buys bonds, increasing the amount of high power money (monetary base)
supplied

Remember that the supply of money is given by Ms = H / [c + θ (1 – c)], where H is the monetary base, and θ
is the required reserve ratio.
● The central bank can undertake open market operations to increase the supply of base money H. In
open market operations, the CB exchanges base money for government bonds. This means that the
total supply of money increases by an amount depending on the money multiplier.
● Relaxing the reserve requirements, i.e. decreasing θ. Banks would hold a smaller fraction of their
deposits as reserves in the central bank will limit their deposit creation, increasing the money
multiplier, and thus the supply of money.

5. Consider that Pedro Farreró holds 25 euros in his wallet, 10.000 euros in German public debt
(maturity 10 years), 3.000 euros in other financial assets with a maturity of 1 year, 6.000 euros in his
bank (which can be used immediately), 60 euros at home, and a house that has a value of 100.000
euros. How much money does he have?

Note that money is an asset that can be used for transactions. Here, these are:

● 25 euros in Pedro’s wallet

● 6.000 euros in Pedro’s bank account
● 60 euros at home
So money = 27 + 6000 + 60 = 6085 euros

6. Assume that the public holds all money in cash, the ratio of reserves to deposits is 0.05, the demand
for money is Md = €Y (0.25 – i), nominal GDP is 1000€, the monetary base is 200€.

a/ What is the equilibrium interest rate?

If consumers and firms hold all the money in cash (c=1), there are no deposits, and M = H = 200€. Then in
equilibrium Ms = Md --> 200 = 1000 (0.25 – i) --> 0.20 = 0.25 – i --> i = 0.05 .

b/ What is the new equilibrium interest rate if the ratio of reserves to deposits increases to 0.5?

Since there are no deposits, the money multiplier remains 1 (the reserve ratio is irrelevant), so the interest
rate does not change.

c/ Assume that the public holds a fraction c = 0.6 of cash now and the reserves ratio remains at 0,05. What
is the total money supply?

Being the money multiplier

1
( c + q (1 - c ) )
1
Ms = mm* Hs = * 200 = 1,61 * 200 = 322,6 € (is the overall money supply)
0,6+0,05(1−0,6)
Universitat Pompeu Fabra
Introduction to Macroeconomics 2022-23 Prof: Teresa Garcia-Milà
Maria Gundín
Alberto Martín

Problem set 3

Question 1. Read the article, “Most stimulus payments were saved or applied to debt”, published by
NBER on October 10, 2020.

a) Explain how the stimulus provided by CARES would be considered in our goods market
model, how it would shift the IS curve and the equilibrium in the economy in an IS-LM model
It is a subsidy, that decreases net taxes, and thus reduces the value of T. A reduction in T
increase disposable income of consumers, increases consumption and shits the IS curve to the
right. The new equilibrium, given a flat LM, will be at a higher value than before the subsidy,
and the size of the increase in income will be related to the multiplier.

b) The article states: Survey data on household behavior suggest that nearly 60 percent of the
stimulus spending went to pay off debt or was saved.

According to this statement, and assuming c0 is unchanged, what would be the marginal
propensity to consume and to save?

The marginal propensity to consume on this subsidy has been empirically observed to be only
40%, and the marginal propensity to save 60%

c) The article continues: The averages mask considerable variation among households. Some 20
percent saved virtually all of their stimulus check; another 40 percent spent nearly all of it.
Roughly 20 percent used most of their federal payment to reduce their debts.

The above paragraph describes consumers with different consumption behavior. If we

summarize those differences in terms of the marginal propensity to consume, identify that
marginal propensity to consume for the three types referred to in the paragraph

For the first group the marginal propensity to consume on the subsidy received is near zero. For
the second group is near to one. And for the third group the propensity to consume is also near
zero as they save those resources and help reducing the negative accumulated savings of
previous periods (debt).

d) Also, in the article you can read: The researchers discovered a number of interesting patterns.
Larger households leaned toward spending most of the money. Seniors tended to pay down debt
while younger and more educated households were more likely to save the payments. Those
who were out of the labor force or who lived with parents were more likely to spend. African
Americans were more likely to use most of their stimulus money to pay down debts, and
Hispanics were more likely to spend.

Comment these different patterns and reconcile them with the consumption behavioral equation
presented in our model.

2
 The article confirms that agents in the economy behave differently accordingly to the moment
in their life cycle, economic conditions, and cultural background. This is compatible with the
consumption behavioral equation of our model as it represents a representative consumer,
which is in fact a weighted average of the different types of consumers in real economies.

e) What does the article conclude respect to the effectiveness on stimulating the economy of
transfer packages like CARES?

It is summarized in this paragraph: The researchers say this suggests that there is a bound on
how much stimulus can be generated through direct transfers to households, and that in the
face of large crises, decision-makers may want to consider a broad range of policies targeting
aggregate demand, with direct transfers being only a part of the fiscal policy response.

f) Suppose based on the article, that the population of the economy is divided into two groups of
equal size: one half of the population with c1 low, and one half of the population with c1 high.

1. Compute the multiplier of a government subsidy S that is distributed equally among the
entire population.

On average the marginal propensity to consume will be

c1 avg = (c1 low + c1 high)/2
And thus we can consider that the marginal propensity to consume of the representative
consumer is c1 avg and thus the multiplier for S in our simple goods market model will
be [1/(1-c1 avg)] and the impact on output will be S*[1/(1-c1 avg)]

2. Suppose instead that the subsidy S is distributed only among the population with c1 high.
What is the multiplier of the subsidy in this case? Is it higher or lower than before?
Explain.

The relevant marginal propensity to consume is now only c1 high. The multiplier is thus
[1/(1-c1 high)], larger than before. The impact on the subsidy in the economy is larger,
and equal to: S*[1/(1-c1 high)]

As now only those with high propensity to consume receive the subsidy, the proportion
of the subsidy consumed is larger than before, and thus the impact on aggregate demand,
and its multiplier effect on output, is larger.

Question 2. Use an IS-LM diagram to show the effects on output of a decrease in taxes T. Can you tell
what happens to investment? Why? Consider the two cases: a) upward slopping LM function;
b) flat LM function.
* Compare the two cases, a) and b), explain why the LM curves are different, and what differences you
see in the impact of a decrease in T on the economy.
IS: Y = C + G + I + X – IM
Decrease in T will shift the IS curve to the right (for every level of interest rate we have higher output).
As you can see in the figure.

2
In the new equilibrium we have higher output and higher interest rate. Investment has a positive relation
with output and a negative relation with interest rate I(Y,i). Because both have increased by a decrease in
T, we have 2 opposing effects on Investment and the outcome is ambiguous.

Now let us consider a flat LM curve, we have:

IS: Y = C + G + I + X – IM
LM: i = î (blue line)

A decrease in T will shift the IS curve to the right. In this case the interest rate is chosen fixed by the
central bank, so we do not see any change in the interest rate as a result of the fiscal expansion. In the

2
new equilibrium we have higher output. Investment has a positive relation with output I(Y,i). So, we will
see an increase in investment due to the fiscal expansion, as output increases and interest rate does not
change.
To keep the interest rate constant the Central Bank will increase Money Supply, to meet the new money
demand that will have increased as a result of the increase in output.
Please note: compared to the upward sloping LM the increase in output is higher as it is shown in the
graph by the dashed upward LM. This is because the interest rate does not increase, so the negative impact
on investment is not present anymore.

Question 3. Consider an economy represented by the following IS-LM model:

C=150+0. 5(Y-T)
I= 250+0.25Y -500i
G=400
T=100
î = 0.15
a. Derive the IS equation. Obtain an expression of Y as a function of all other variables. Represent the
equation in a graph. Add in the graph the LM equation. Label the axis correctly.

Y = C + I + G = 150+0. 5(Y-100) + 250+0.25Y -500i + 400

Y = ( 1/( 1–0.25–0.5)) (750-500i)= 4(750-500 i)

IS: Y = 4 (750-500 i)
LM: i=0.15

b. Find the equilibrium output in this economy and show it in the graph

2
 Y = 4 (750-500 x 0.15); Y*=2700

c. What is the level of real money supply when the interest rate is 15% and the economy is at the
output of equilibrium? To answer this question, consider that the demand for real money balances is
equal to 2Y-8000i, and that the economy is in equilibrium in the money market as well as in the
goods market.

M/P* = 2 Y* – 8000 i* = 2*2700 – 8000*0,15 = 4200

d. Solve for the equilibrium values of C, I.

Y*=4 (750- 500 x 0.15)= 2700

i* = 0.15
C*= 150+0. 5(Y*-T)= 150 + 0.5 (2700 – 100) = 1450
I* = 250+0.25Y* -500i* = 250 + 0.25(2700) – 500*0.15 = 850

C* + I* + G = 1450 + 850 + 400 = 2700 = Y*

e. Now suppose that the central bank increases the interest rate to 20%. How does this change the LM
curve? Solve for Y, I and C, and describe in words the effects of this policy. What is the new
equilibrium value for the real money supply M/P? Use a graph to go along with your explanation.

Y* = 4 (750-500 i*) = 4 ( 750 – 500*0,20 ) = 2600

C*= 150+0. 5(Y*-T)= 150 + 0.5 (2600 – 100) = 1400
I* = 250+0.25Y* -500i* = 250 +0.25(2600) – 500*0,20 = 800

As we can see the effect of a contractionary monetary policy is a reduction in output, consumption, and
investment.
To reach the new interest rate, the Central Bank has decreased the money supply, selling bonds through
OMO. The real money supply is:
M/P* = 2Y*– 8000i* = 2*2600 – 8000*0.20 = 3600

2
f. With this new monetary policy î = 0.20, design a fiscal policy that by changing G would take the
economy back to the original equilibrium output obtained in question c above. Will the values for C
and I be the same now as in question d.?

Y* = 4 (350-500 i* + Gnew )= 2700

4 (350-500*0.20 + Gnew )= 2700

Gnew = 2700/4-(350-500*0.20) = 425. The government needs to follow an expansionary fiscal policy,
increasing G by 25 to recover the initial level of Y.

C*new = 150+0. 5(Y*-T) = 150 + 0.5(2700 – 100) = 1450. Same as in part d. as it is not function of i

I*new = 250+0.25Y*-500i* = 250 + 0.25(2700) – 500*0.20 = 825

I* is smaller than in part c. as the increase in interest rate reduces Investment for the same output. Note
that the decrease in I is equal to the increase in G.

g. With this same monetary policy î = 0.20, design now a fiscal policy that by changing T would take
the economy back to the original equilibrium output obtained in question d above. What will the
values for C and I?

Y* = 4 (800 -500 i* - 0.5 Tnew ) = 4(700 - 0.5 Tnew) = 2700

Tnew = 50. If the government wants to follow an expansionary fiscal policy similar in impact to the one
in question e, but by cutting taxes instead of increasing government expenditures, the reduction of taxes
needs to be larger to impact autonomous spending by the same amount (i.e. 25 in part f.). That is

2
because the impact on autonomous spending of a reduction in taxes is affected by the marginal
propensity to consume, 0.5 in this economy. Thus, to increase autonomous spending by 25, the
government needs to reduce taxes by 50. In this economy taxes would have to decrease to 50.
Consumption will increase to 1475.
C*new = 150+0. 5(Y*- Tnew) = 150 + 0.5(2700 – 50) = 1475.
Although income has not changed, disposable income has increased because taxes have decreased from
100 to 50. Investment would be identical to the one in question e, as the equilibrium income and interest
rates are the same.

h. Now return to the initial situation with the policy interest rate at 15%. Suppose that government
spending increases to 600 and taxes adjust so that the government has a balanced budget. How does
this change the IS curve? Solve for Y, I and C, and describe in words the effects of this policy.

G = T = 600. i*= î = 0.15

Y = C + I + G = 150+0. 5(Y-600) + 250+0.25Y -500i + 600

IS: Y=4 (700 – 500i)

LM: i=0.15

Y*= 2500
i*=0.15
Autonomous spending decreases by 50 units (from 750 to 700), a result of an increase of 200 in G and
an increase in taxes of 500 that multiplied by the marginal propensity to consume, 0.5 decreases
autonomous spending by 250.

There is therefore a decrease in output of 200 units (from 2700 to 2500).

C* = 150 + 0. 5(Y*-T) = 150+0. 5(2500 - 600) =1100

I* = 250 + 0.25Y* - 500i* = 250 + 0.25 (2500) - 500*0.15 = 800
G = 600

C* + I* + G = 2500

There is a reduction in both C* and I*. The decrease in Y* and the increase in T reduces C*.
Investment is reduced due to the reduction in Y*.

i. Assume now that the investment behavior of this economy changes, and investment is more
sensitive to changes in the interest rate. Specifically assume I = 250 + 0.25Y - 600i.
How does that change affect the IS function, will it be steeper or flatter? Make a graph showing the change
in the IS.
Calculate the new equilibrium with the original monetary policy at an interest rate = 0.15. Obtain output,
consumption, and investment. Comment on the differences with the original equilibrium you found in
question a.

2
Y* = 150 + 0.5(Y-T) + 250 + 0.25Y – 600i + 400
Y* = 4(750 – 600i) = 2640
C = 150 + 0.5(2640 – 100) = 1420
I = 250 + 0.25*2640 – 600*0.15 = 820

The IS curve will become flatter, as output is now more sensitive to the interest rate (through
investment). With the same interest rate i = 0.15, investment and thus equilibrium output are lower. Due
to lower output, equilibrium, consumption is also lower than in question a.

2
Universitat Pompeu Fabra
Introduction to Macroeconomics 2021‐22
Teresa Garcia‐Milà, María Gundín, Alberto Martín
Problem set 4

SECTION A

Watch the video “How does raising interest rates control inflation”
https://www.youtube.com/watch?v=R8VBRCs2jTU&t=5s. Answer the following
questions based on the information provided by the video.

1. Explain how lenders and borrowers view differently the movements in the
interest rate, and how an increase of the interest rate affects them differently.
If you are a borrower (i.e. you are taking a loan), you want the interest rate as
low as possible so that you pay back relatively less eventually. On the other
hand, if you are a lender (i.e. you have savings), than a high interest rate means
you can earn more money on your savings.
2. Summarize how the rise of interest rates by the central banks transfers to
raises of interest rates of commercial banks to their customers.
A central banks acts like a bank for banks. Just as people with a savings
account, commercial banks also earn interest when they leave money
(reserves) with the central bank. Commercial banks can deposit their excess
reserves at the central bank to earn interest rate on it. When central banks
raise interest rates, a commercial bank will earn more on its reserves. It might
make more from keeping their money in a central bank than lending it out. So,
if they do lend it out, they will raise their interest rates, to make it worthwhile
for them.
3. The video explains that mortgages can be at fixed interest rates or at variable
interest rates. Explain the difference and how in each case the impact in the
economy is a decrease in aggregate demand.
If you’ve got a variable‐rate mortgage, where the interest rate that you pay is
linked to the central bank interest rate, then higher interest rates mean that
your monthly payment on your debt will increase. Thus, essentially
immediately the higher rate will translate into less cash to spend on other
things.
People with fixed‐rate mortgages are protected against the direct effect of an
interest rate rise. However, they will still feel the indirect impact. High interest
rates mean that mortgages become more expensive, thus new home buyers
will have less money to spend. Then, we expect house prices to fall. That will
make everyone who owns a home (including those with a fixed‐rate mortgage)
feel poorer – and therefore they might spend less.
4. Explain the mechanisms through which an increase in interest rates affects
firm’s decisions.
When interest rates rise, then businesses will find it more expensive to borrow
and invest. That generally means less economic activity, with potentially less
jobs created, and lower salaries for employees.
5. Why is, according to this video, high inflation for a long time a problem for the
economy?
 Higher prices mean employees will need higher wages, pushing up costs for
businesses. That could drive up prices further, potentially leading to an upward
spiral of wages and prices. It might also increase inflation expectations, making
bringing down inflation much more costly.

SECTION B

Question 1
Consider the following IS‐LM model:
IS: Y=C(Y‐T)+I(Y,r+x)+G
LM: r=r

Where r is the real interest rate that results from the federal funds nominal rate, i, and
expected inflation. Unless otherwise stated, we consider expected inflation to be
constant. Assume that the rate at which firms can borrow is much higher than the
federal funds rate, equivalently that the risk premium, x, in the IS equation is high.

a1) Suppose that the government takes action to improve the solvency of the financial
system. If the government’s action is successful and banks become more willing to
lend ‐ both to one another and to non‐financial firms – what is likely to happen to the
premium?
The risk premium, x, will decline. If the government succeeds with its plan to improve
the solvency of the financial system and banks are more willing to lend both to one
another and to non‐financial firms, this means they will be willing to lend at a lower
interest rate: 𝑖 ↓ 𝑟 𝑥 ↓ (if x declines, 𝑖 declines).

a2) What will happen to the IS-LM diagram? Show the changes in a diagram, and make
sure you label the axis appropriately.
The decrease of x will shift the IS curve to the right, as investment will increase for any
given level of output Y/for any given level of r. Hence, assuming no change in
monetary policy, output increases
b1) Faced with a zero nominal interest rate, suppose the FED decides to purchase
bonds directly to facilitate the flow of credit in the financial markets. This policy is
called quantitative easing. If quantitative easing is successful, so that it becomes easier
for financial and non‐financial firms to obtain credit, what is likely to happen to the
premium?

If the Federal Reserve buys bonds in the market it is increasing the demand for bonds,
which raises its price and lowers the associated interest rate (via x, since 𝑖 is in the zero
lower bound and cannot decrease).
For the same level of r, the interest rate that firms face now is lower, which means the
premium x decreases ( 𝑖 ↓ 𝑟 𝑥 ↓)

b2) What effect will this have on the IS‐LM diagram?

The same as in a1 (assuming no change in expected inflation).

b3) If quantitative easing has some effect, is it true that the FED has no policy options
to stimulate the economy when the federal fund rate is zero?
No, it’s false as shown in the last plot. By buying securities directly, the FED increases
the price of private bonds and decreases therefore the interest rate firms face to
borrow, stimulating investment. In our model this is represented by a reduction in the
risk premium, x, faced by firms. So, the federal funds rate remains at zero, but the
interest rate faced by borrowers decreases, and has an expansionary effect on the
economy.

Question 2
Consider the following IS‐LM model:
C= 200+0. 25(Y‐T)
I= 400+0.25Y ‐800(r+x)
G=800
T=200
r= r =0.05
where r is the real interest rate that results from the nominal policy rate decided by
the central bank and expected inflation.
We assume that expected inflation is constant and equal to 0.03.
Assume that the risk premium is x=0.07
a) Derive the IS relation.
Y=C+I+G
Y= 200+0.25(Y‐200) + 400 + 0.25Y –800(r+0.07) + 800
Y = 2588 ‐1600r or r = (‐1/1600)Y + (2588/1600)
b) What is the level of real money supply when the real interest rate is 5% and the
expected inflation is 3%? To answer this question, use the following equilibrium
equation in the money market:
M/P=4Y‐5000i
First, we have that when r=0.05, Y= 2508 in equilibrium.
Also, the Fischer equation tells us that i=r+ Πe, then i=0.08 (since r=0.05 and Πe=0.03).
Hence: M/P = 4*(2508) – 5000*(0.08) =10.032 ‐ 400
M/P=9632
c) Solve for the equilibrium values of C, I and Y.
Y=2508 in equilibrium, so…
C= 200+0.25(Y‐200) = 200 + 0.25(2540‐200)
C= 777
I = 400 + 0.25Y –800(r+0.05) = 400 + 0.25*2508 –800(0.05+0.07)
I = 931

d) Now suppose that the central bank cuts the policy nominal rate so that, with
constant expected inflation, the resulting real interest rate is 2%. How does this new
policy change the LM curve? Solve for the equilibrium values of Y, I and C, and describe
in words the effects of this policy. What is the new equilibrium for the real money
supply M/P? Show in a graph the changes.
Before we got that:
Y= 2588‐1600r
As now r=0.02, then Y= 2556 in equilibrium.
C=200+0.25(Y‐200) = 200 + 0.25(2556‐200)
C=789
I = 400 + 0.25Y –800(r+0.07) = 400 + 0.25*2556 –800(0.02+0.07)
I = 967

The decrease in the interest rates (both nominal and real (since expected inflation
does not change)) leads to a higher output first because of a higher investment. As the
interest rate is lower, there are more incentives to invest. As output increases,
disposable income increases and consumption increases, leading to an even larger
increase in output.
The new equilibrium in the money market will be:
The Fisher equation tells us that i=r+ Πe, then i=0.02+0.03=0.05
Hence: M/P = 4*(2556) – 5000*(0.05) =10.224 ‐ 250, M/P=9974
In words, a decrease in interest implies necessarily that, in equilibrium, money supply
increases. It is an expansionary monetary policy.
r

LM0 r=0.05

LM1 r=0.02

IS0

Y0 Y1 Y
e) Now return to the initial situation with the policy nominal rate such that, with
constant inflation, the real interest rate is equal to 5%. Suppose that economic
confidence has worsened, solvency of the financial system has deteriorated and as a
result the risk premium has increased to 10%. How does this affect the IS and the LM
curves? Solve for the equilibrium values of Y, I and C, and compare with the initial
equilibrium values. Explain the differences. Show in a graph the changes.

An increase in the risk premium x shifts to the left the IS equation, since for the same r,
companies have less incentives to invest as it is “more expensive” for them to borrow
money. Investment, consumption and output will be lower.
In the LM there is no change.
Y=C+I+G
Y= 200+0.25(Y‐200) + 400 + 0.25Y –800(r+0.10) + 800
Y = 2540 ‐1600r or r = (‐1/1600)Y + (2540/1600)
With r=0.05, Y= 2460
C= 200+0.25(Y‐200) = 200 + 0.25(2460‐200)
C= 765
I = 400 + 0.25Y –800(r+0.10) = 400 + 0.25*2460 –800(0.05+0.10)
I = 895
We obtain a lower output, consumption and investment than in the original situation.

f) Suppose the government carries on an expansionary fiscal policy to try to counteract

the decrease in the output of equilibrium after the worsening of the confidence
conditions in the economy. Assume G is increased up to 1200. Calculate the new
equilibrium, values for Y, C and I, compared it with the one obtained in e). What
impact will this government action have on the government budget?

An increase in government spending shifts the IS curve to the right, since for the same
r, output is higher. Government spending generates higher income, which allows
households to consume more and increases business investment.

In the LM there is no change.

Y=C+I+G
Y= 200+0.25(Y‐200) + 400 + 0.25Y –800(r+0.10) + 1200
Y = 3340 ‐1600r or r = (‐1/1600)Y + (3340/1600)
With r=0.05, Y= 3260
C= 200+0.25(Y‐200) = 200 + 0.25(3260‐200)
C= 965
I = 400 + 0.25Y –800(r+0.10) = 400 + 0.25*3260 –800(0.05+0.10)
I = 1095

Compared to the results obtained in e), we have higher output, consumption and
investments. One extra euro of government spending generate more than one euro in
output, as the fiscal multiplier is 2. Therefore, as government spending
. .
increased by 400, output is higher by 800.

The government budget is given as T‐G. Assuming that taxes remain unchanged at 𝑇
200, we have that the budget deficit has deteriorated by 400:
𝑇 𝐺 200 800 600
𝑇 𝐺 200 1200 1000
Question 3
Consider the following IS‐LM model:
IS: Y=C(Y‐T) + I(Y,r+x)+G. Asume: C(Y‐T)= c0 + c1(Y‐T); I(Y, r+x) = b0 + b1Y – b2(r+x)
LM: r=r
Where r is the real interest rate that results from the federal funds nominal rate, i, and
expected inflation (which we assume to be equal to zero for this exercise).
Suppose that T=T0+t(B*r), where t is a positive coefficient and B denotes the stock of
public debt. This is a simple way to capture the fact that tax revenues must increase
alongside debt payments in order for the government to be able to service its public
debt.

a1) How does the shape of the IS curve depend on the stock of debt B? Draw two IS
curves for a high and a low level of debt, respectively.
We can obtain the IS curve by imposing the equilibrium condition in goods market:
Y= c0 + c1(Y‐T0‐t(B*r)) + b0 + b1Y – b2(r+x) + G
Now, using that i=r if expected inflation is 0, and solving for i yields:
i= ∗𝑌 ∗ (c0 –c1T0 + b0‐ b2x+G)
∗ ∗ ∗ ∗

We can see that the stock of public debt affects both the slope and the intersection on
the vertical axis of the IS curve. A higher value of B will lead to a flatter curve with a
lower intersection (Assume BL<BH). The intersection on the horizontal axis, when i=0 is
for the same Y as at zero interest rate there is no cost for the debt.

b1) Suppose two economies with different levels of public debt, as you have
considered in the previous question. Suppose that the central bank raises the interest
rate. What are the effects of an interest rate hike in both economies? How do these
effects differ? Explain.
Using the graph in the previous question, we can see that an interest rate hike will lead
to a higher contraction in the country with higher stock debt.
When interest rates increase, the bigger the stock of debt, the bigger will be the
increase in T, which leads to a higher decline in output (the government will need to
collect more taxes as it will be facing higher interest payments).

c1) Can you use this simple example to think about the euro area economy? Explain.
This is a good example for the euro area, where there are some countries with higher
stocks of debt (s.a.: Spain, Italy, Greece or Portugal) than others (s.a. Germany or
Netherlands) and all are under the same monetary policy from the ECB. Using this
model, the latter countries are “better prepared” to face the rate hikes that the ECB is
recently applying.

Debt as a percentage of GDP of some euro area countries:

2020 2021 2022
Austria 83.3 82.9 78.5
Belgium 112.8 108.4 103.9
Estonia 18.6 17.6 18.3
Finland 69.0 66.2 66.7
France 114.7 112.6 111.8
Germany 68.0 69.6 71.1
Greece 212.4 199.4 177.6
Ireland 58.4 55.3 47.0
Italy 155.3 150.9 147.2
Netherlands 54.6 52.3 48.3
Portugal 135.2 127.4 114.7
Slovak Republic 59.7 63.1 60.5
Slovenia 79.6 74.4 69.5
Spain 120.0 118.6 113.6

IMF, World Economic Outlook Database, October 2022

Assuming the previously described model, rate hikes from the ECB will have a more
negative impact on output in countries with higher stock of public debt than in
countries with low values of public debt.
Universitat Pompeu Fabra Introduction to Macroeconomics 2022‐23

Problem set 5

Question 1
Consider the price‐setting relation (in the simple case when the marginal product is constant and
equal to one):

𝑊 1
𝑃 1 𝑚

where m is the mark‐up over the marginal cost that firms can charge when they have market power
in the goods markets.

a. Explain why a fall in the mark‐up m implies that the real wage is higher?

Reminder – Derivation of price setting relationship

Firms set prices following the rule “𝑝𝑟𝑖𝑐𝑒 𝑚𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑐𝑜𝑠𝑡 1 𝑚𝑎𝑟𝑘𝑢𝑝 ”. Mathematically:

𝑃 𝑊∗ 1 𝑚 (1),

Where marginal cost = 𝑊.

The marginal product of labor tells us how much output changes when we change labor.
Mathematically: 𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙 𝑃𝑟𝑜𝑑𝑢𝑐𝑡 𝑜𝑓 𝐿𝑎𝑏𝑜𝑟 ≡ 𝐴, and 𝐴 1
(warning: we won’t always assume 𝐴 1).

To derive the price‐setting relationship, just rewrite equation (1) as:

𝑊 1
2
𝑃 1 𝑚

Solution
Now, for a given nominal wage, if the market power of firms decreases, firms will decrease their
prices 𝑃 ↓ W ∗ 1 𝑚 ↓ . This means that, for any given nominal wage, the real wage must
increase

𝑅𝑒𝑎𝑙 𝑊𝑎𝑔𝑒 ↑ ≡
↓ ↓

When losing market power, the mark up is reduced, therefore, prices fall. And, because in the short
term there are sticky wages, real wage must increase consequently.

b. Draw this price‐setting relation on the (u, W/P) graph, where u is the unemployment rate and
W/P is the real wage. Graphically show how the real wage changes when there is a decrease in the
markup.
Solution
Since in the price setting relation; is NOT a function of 𝑢, it is just a horizontal line at .
If markup decreases from 𝑚 to 𝑚 , the price setting line shifts up leading to an increase in real
wage . Graphically:

c. Now draw the wage‐setting relation and explain this relationship.

Wage setting relation, where u=unemployment rate and 𝑧 ≡ all other factors that affect wage
determination from unemployment benefits to the form of collective bargaining (labour market):

The wage setting curve implies a negative relationship between real wages, W/P, and the
unemployment rate u. The higher the unemployment rate the weaker the bargaining position of
the workers and therefore the lower the real wage will be. As an aside, it should also be noted that
an increase (decrease) in 𝑧 would shift the WS curve to the right (left).

d. Assume that, due to technological change, there is an increase in market power (e.g. Amazon).
What will it happen to the natural rate of unemployment?
Reminder – Equilibrium in the labour market
We have two schedules, the price‐setting relationship ( ) and the wage‐setting relationship
( 𝐹 𝑢, 𝑧 ). At their intersection, both firms are setting their prices optimally AND workers are
setting their wages optimally, which means that, at the point ( , 𝑢 ), both firms and workers are
optimizing (i.e., the labor market is in equilibrium). We call 𝑢 the “natural rate of unemployment”
(mathematically, 𝑢 ≡ 𝑢 ).

Solution
An increase in market power is modelled as 𝑚 ↑ (from 𝑚 to 𝑚’, where 𝑚 𝑚’), which means that
the price‐setting relationship shifts down:

In the new labour market equilibrium, (i) the real wage has decreased and (ii) unemployment has
increased. Intuitively: if Amazon gains market power, then it can set higher prices in equilibrium for
a given nominal wage, or, in other words, it will pay lower real wages. Since the real wage paid by
firms is lower, for a given level of 𝑧, workers have to accept a higher level of unemployment for the
wage‐setting relationship to hold ⟹ ↓ 𝐹 𝑢 ↑, 𝑧 ). This is why 𝑢 ↑ (from 𝑢 to 𝑢 ′).

Intuition behind this:

Firm gains market power and thus can rise prices, and because nominal wages are sticky, real wages
decreases. In equilibrium, workers accept these lower real wages only if their bargaining power is
diminished. This is achieved through an increase in unemployment as the economy moves along
the wage‐setting curve.

e. Suppose that the parliament approves a rise in the minimum wage to counter this effect. Show
graphically the new equilibrium.

In this model, a rise in the minimum wage is modelled as 𝑧 ↑ (recall: 𝑧 captures all the factors that
affect the real wage other than unemployment). Thus, this change affects the WS schedule, and
graphically this is represented by a shift upwards of the WS (meaning: “for a given level of
unemployment, demanded by workers is larger because their bargaining power is enhanced by a
larger real minimum wage”.)
Since the PS relationship is not affected, we have:

NOTE: This graph is a continuation of the one in exercise 1d. It just adds the new shift of the WS
schedule.

Thus, when the labour market is in the new equilibrium, the natural unemployment rate has
increased even further (from 𝑢 ′ to 𝑢 ′′), and the wage has remained the same (still at
).

Intuition behind thi

By increasing the minimum wage, the government strengthens the bargaining position of workers
and – all else equal – makes them demand a higher wage in equilibrium (think, for instance, that
their “reference” point has shifted). Since the situation of firms has not changed, however, they are
only willing to pay the same wage as before. How can the same wage be an equilibrium given the
higher bargaining power of workers? The answer is that unemployment increases, weakening the
bargaining power of workers and offsetting the effect of a higher minimum wage.

f. What happens to the real wage and the natural rate of unemployment?

We’ve seen, graphically, that the real wage does not change in equilibrium, but the unemployment
rate increases.
Intuitively, an increase in 𝑧 ↑ is an increase in the bargaining power of workers (in this case, because
workers feel more “empowered” to ask for higher wages when the minimum wage is raised by the
government), who will push for higher wages. Firms, on their side, have their prices set as function
of the wages already bargained with unions and as a function of their market power. That means
that the real wage is set (by setting prices, firms are the ones that ultimately set the real wage in
this model). To achieve a new equilibrium, it is necessary that the bargaining power of workers goes
back to the initial level for them to accept the original real wage. That is attained if the
unemployment rate is higher than before the 𝑧 had increased.

g. According to this model, what is a better policy than the minimum wage in order to raise real
wages?

As said, after negotiating 𝑊 (nominal wage) with unions, firms set their prices depending on their
markup 𝑚. In other words, 𝑚 is the ultimate factor affecting the real wage in this model.
Now, if the government wants to achieve a larger real wage in equilibrium, it has to design antitrust
policies intended at lowering the markup 𝑚: reduce collusion between firms, eliminate monopolies,
foster competition, etc. These policies will lower the markup, which means that the price level 𝑃
will be lowered for a given nominal wage 𝑊 bargained with the unions (recall the PS relationship:
𝑃 ↓ 𝑊⏞ ∗ 1 𝑚 ↓ ), which means that the real wage will increase in equilibrium.

Question 2

Consider the following wage and price setting equations in the country of Mediterranea:

‐ wage‐setting relation (WS): 0.2 𝑧– 𝑢

‐ price‐setting relation (PS):

a. Suppose Mediterranea signs a free‐trade agreement with Asia, which reduces the market power
of domestic firms. What is the effect on equilibrium wages and unemployment? Explain.

In this model, a reduction of market power in domestic firms is modelled as 𝑚 ↓ (recall: 𝑚 captures
the mark‐up over the marginal cost, when firms have market power, m is positive). From the price‐
setting relation, if m decreases prices decrease: 𝑃 ↓ 𝑊⏞ ∗ 1 𝑚 ↓ . This means that, for any
⏞
given nominal wage, the real wage must increase (𝑅𝑒𝑎𝑙 𝑊𝑎𝑔𝑒 ↑≡ ). Graphically, this is
↓
represented by a shift upwards of the price‐setting relationship (from 𝑚 to 𝑚’, where 𝑚 𝑚′).
In the new labor market equilibrium, (i) the real wage has increased and (ii) unemployment has
decreased. Intuitively: if firms loose market power because there are more firms competing
internationally, then they will have to set lower prices in order to be competitive. In equilibrium,
for a given nominal wage, if prices decrease the real wage will increase. Since the real wage paid by
firms is higher in equilibrium, for a given level of 𝑧 workers must have seen an increase in their
bargaining power in equilibrium, which means that unemployment must have decreased for the
wage‐setting relationship to hold ⟹ ↑ 𝐹 𝑢 ↓, 𝑧 ). This is why 𝑢 ↓ (from 𝑢 to 𝑢 ′).

Same Intuition Exercise 1.d):

Due to the lower market power, firms are willing to reduce prices for any given level of the wage,
giving rise to an increase in the real wage. This higher real wage is consistent with equilibrium only
if workers have a higher bargaining power, i.e., they can demand a higher wage because it is less
costly for them to look for another job when unemployment is very low.

IMPORTANT NOTE: A stricter graphical representation here would have been plotting the WS
schedule as a straight downward sloping line (we’re given a linear WS schedule in the exercise with
a slope of ‐1). The curve plotted, however, suffices to illustrate the effects of 𝑚 ↓.

b. Once the trade agreement is signed, it becomes easier for firms to shift production abroad.
Because of this, the bargaining power of workers decline, which we can capture by a decline in z.
What is the effect on equilibrium wages and unemployment? Explain.

A fall in workers’ bargaining power is captured through a decline in z and thus a downward shift of
the wage‐setting curve (i.e., for any given level of unemployment, workers are willing to work for a
lower wage). This means that the wages paid by firms – which are only a function of productivity
and their markup – can only be consistent with equilibrium if the decline in workers’ market power
is offset through a lower level of unemployment.
The bottom line when both a) and b) are considered is that, viewed through the lens of this exercise,
a trade agreement raises real wages and reduces unemployment.

c. Given your previous answers, are workers in Mediterranea in favor of the trade agreement or
not? Explain your reasoning.

They are in favor of the trade agreement, as it reduces unemployment and increases real wages.

Question 3

Consider a labor market that works according to the wage‐ and price‐ setting equations as seen in
class. Suppose that the government mandates that employers provide higher benefits to workers,
in a manner proportional to their wage. Specifically, besides paying the wage W, the employer must
supply an additional 𝜌 𝑊.

What happens to unemployment and the real wage as a consequence of this regulation? Show your
answer graphically and provide an intuition.

The employer must supply an additional benefit of 𝜌 𝑊. Assuming that firms have access to a
production function given by 𝑌 𝐴𝑁 i.e., 𝑀𝑃𝐿𝑎𝑏𝑜𝑢𝑟 𝐴 , this will increase the marginal cost
∗
of producing for the firm to , which means that the new price setting equation will be 𝑃
∗
∗ 1 𝑚 ⟺𝑃 ∗ 1 𝜌 1 𝑚 . Obviously, for 𝜌 0 prices increase 𝑃 ↑ ∗
1 𝜌 1 𝑚 .
Rewriting, we can arrive to the expression for the real wage:

𝑊 𝐴
𝑃 1 𝜌 1 𝑚

Now, for any given nominal wage (𝑊), marginal product of labor (𝐴), and mark‐up (𝑚), the real
⏞
wage must decrease (𝑅𝑒𝑎𝑙 𝑊𝑎𝑔𝑒 ↓≡ ). Graphically, this is represented by a shift downwards
↑
of the price‐setting relationship (from 𝑃𝑆 to 𝑃𝑆’).
In the new labor market equilibrium, (i) the real wage has decreased and (ii) unemployment has
increased. Intuitively: if there are higher benefits to workers, the marginal cost increases, which in
turn means that firms will set higher prices. In equilibrium, for a given nominal wage, if prices
increase the real wage will decrease. Since the real wage paid by firms is lower, for a given level of
𝑧, workers have to accept a higher level of unemployment for the wage‐setting relationship to hold
⟹ ↓ 𝐹 𝑢 ↑, 𝑧 ). This is why 𝑢 ↑ (from 𝑢 to 𝑢 ′).

Similar Intuition to 1.e):

If Government rises minimum benefits to workers, the marginal cost of production of the firm
increases, and because the mark up hasn’t changed at all the firm will translate all increase to prices
and the real wage falls. This lower real wage can only be consistent with equilibrium if the
bargaining power of workers is weakened, which happens here through an increase in the
equilibrium level of unemployment.

Note that, in this answer, we have assumed that the higher benefits (amenities) are not valued by
the workers as the monetary wage W. Hence, the WS equation was not affected. Instead, if these
higher benefits were valued by workers in the same manner as the wage W, the WS curve would
shift down to ∗ , and equilibrium unemployment would remain unaffected. This is because
now the workers value to benefits in a similar way to wage W and therefore understand that the
real wage implies a total compensation of 1 𝜌 times the real wage.

Question 4

Consider the following equations:

‐ wage‐setting relation (WS): 0.55 𝑧– 𝑢

‐ price‐setting relation (PS):

a. Obtain the expression of the equilibrium rate of unemployment 𝑢 (that is, find the equilibrium
WS=PS).
Mathematically, the level of unemployment such that firms set prices optimally and workers set
nominal wages optimally (i.e., the equilibrium rate of unemployment, aka natural unemployment
rate) can be found by imposing 𝑊𝑆 𝑃𝑆:

𝐴
0.55 𝑧 𝑢 ⟺
1 𝑚
𝐴
𝑢 0.55 𝑧 ≡𝑢
1 𝑚

b. Obtain the expression for the equilibrium real wage .

The real wage is ultimately set by the firms in this model, so we just need to look at the PS
relationship to conclude that…
𝑊 𝐴 𝑊
≡
𝑃 1 𝑚 𝑃

c. If z increases, what happens to 𝑢 , and ?

Mathematically, it is straightforward to see that 𝑢 increases and stays the same:

● 𝑢 ↑ 0.55 𝑧↑
●
.
More formally: 1 (i.e., the natural unemployment rate increases one‐for‐one

with changes in 𝑧), and 0 (i.e., the equilibrium real wage is unaffected by changes
in 𝑧).

NOTE: See exercises 1f‐1g to see how an increase in 𝑧 affects the WS and PS schedules graphically,
and how it affects the equilibrium levels of 𝑢 and . The 𝑢 ↑ that we see in this exercise (6c)
would be represented by 𝑢 ′ increasing to 𝑢 ′′ in the graph in 1f.

d. If m increases, what happens to 𝑢 , and ?

Mathematically, it is straightforward to see that 𝑢 increases and decreases:

● 𝑢 ↑ 0.55 𝑧
↑
● ↓
↑
.
More formally: 0, and 0 (since 𝐴, 𝑚
0).

NOTE: See exercise 1d to see how an increase in 𝑚 affects the WS and PS schedules graphically, and
how it affects the equilibrium levels of 𝑢 and . The 𝑢 ↑ and ↓ that we see in this
exercise (6d) would be represented by 𝑢 increasing to 𝑢 ′ and decreasing to in the graph
in 1d.

e. If there is a decrease in productivity (i.e., if 𝐴 decreases), what happens to 𝑢 and ?

Mathematically, it is straightforward to see that 𝑢 increases and decreases:

↓
● 𝑢 ↑ 0.55 𝑧
↓
● ↓
.
More formally: 0, and 0 (since 𝑚 0).

NOTE: Graphically a decrease in A will shift the PS schedule downwards, causing an increase in the
natural unemployment rate and a decrease in the equilibrium real wage. Intuitively, this happens
because the marginal cost of firms increases (workers are less productive, so producing a marginal
unit of output is more costly for firms), which in turn raises prices and lowers the real wage. Since
firms are paying a lower real wage in equilibrium, workers must have lost some bargaining power
for them to accept a lower wage, which means that the equilibrium rate of unemployment must
have increased.
Universitat Pompeu Fabra
Introduction to Macroeconomics 2022‐23
Teresa Garcia‐Milà, María Gundín, Alberto Martín
Problem set 6
Question 1

Suppose that the Phillips curve is given by

𝜋 𝜋 0.1 2𝑢
Where
𝜋 𝜋

Suppose that inflation in year 𝑡 1 is zero. In year 𝑡, the authorities decide to keep the
unemployment rate at 4% forever.

A. Compute the rate of inflation for years 𝑡, 𝑡 1, and 𝑡 2.

𝜋 𝜋 0 (by assumption)
𝜋 0.1 2 0.04 0.02 2%

𝜋 𝜋 0.02
𝜋 0.02 0.1 2 0.04 0.04 4%

𝜋 𝜋 0.04
𝜋 0.04 0.1 2 0.04 0.04 6%

Note inflation is increasing because unemployment is kept permanently below its natural rate,
which is computed by setting
𝜋 𝜋 0 0.1 2𝑢 → 𝑢 0.05

Now suppose that half the workers have indexed inflation contracts.

B. What is the new equation for the Phillips curve?

Let 𝜆 be the fraction of workers that have inflation indexed contracts. In this example, 𝜆 0.5.

𝜋 𝜆𝜋 1 𝜆 𝜋 0.1 2𝑢 0.5 𝜋 0.5𝜋 0.1 2𝑢

𝜋 𝜋 0.2 4𝑢

C. Re‐compute your answers to A.

𝜋 𝜋 0 (by assumption)

1
𝜋 0.2 4 0.04 0.04 4%

𝜋 𝜋 0.04
𝜋 0.04 0.2 4 0.04 0.08 8%

𝜋 𝜋 0.08
𝜋 0.08 0.2 4 0.04 0.12 12%

D. Based on your previous answers, what is the effect of indexation on the relation between 𝜋 and
𝑢?
Indexed contracts increase the sensitivity of inflation to unemployment, that is, there is a stronger
reaction of inflation to unemployment. In this case, keeping unemployment below its natural rate
leads to an increase in inflation, but that has an immediate effect on wages (through indexed
contracts) and this feeds back into higher inflation.

Question 2

In class, we derived the Phillips curve for the case of 𝐴 1. In this question, we relax this
assumption. It can be shown, in fact, that the Phillips curve for a generic level of productivity is given
by

𝜋 𝜋 µ 𝑧 𝑎 𝛼𝑢

where 𝑎 log 𝐴. Thus, all else equal, inflation is decreasing in productivity: this comes from the
price‐setting equation, where – for given markups and wages – an increase in productivity reduces
production costs and leads to a fall in prices. Note that in the background we are assuming that
higher productivity affects only the price‐setting but not the wage‐setting equation.

Given this Phillips curve, suppose that the economy starts in the medium‐run equilibrium in year 𝑡,
where actual and expected inflation are both equal to 2%. Assume throughout that the Central Bank
follows a policy of adjusting the nominal interest rate to keep the real rate fixed at a given value 𝑟,
and that inflation expectations are fixed at 2%, that is 𝜋 2% for all 𝑡. Suppose that, starting in
period 𝑡 1, the economy experiences a permanent increase in productivity 𝐴.

Using the IS‐LM‐PC model, analyze the effects of the increase in productivity in period 𝑡 1. What
happens to inflation and unemployment?
µ
First, notice that in this example, the natural rate of unemployment is equal to 𝑢 . With
this, we can re‐write the Phillips curve as:

𝜋 𝜋 α 𝑢 𝑢 .

Now, we can re‐write the relationship between unemployment and output:

𝑌 𝐴𝑁 ↔ 𝑌 𝐴 𝐿 𝑈 ↔𝑌 𝐴𝐿 1 𝑢 ↔𝑢 1 such that 𝑢 1

2
We can use this relationship to re‐write the Phillips curve as:
α
𝜋 𝜋 𝑌 𝑌
𝐴𝐿
µ ↑ ↑
An increase in productivity reduces 𝑢 𝑢 ↓ , thereby increasing 𝑌 (𝑢 ↓ 1 ), so
that it pushes the Phillips curve downwards and to the right, as illustrated in the graph below. From
the equation above, we can also see that an increase in productivity reduces the slope of the Phillips
curve (this change in slope is not represented in the graph below).

Question 3

Consider an economy that starts in the medium‐run equilibrium in year 𝑡 , where actual and
expected inflation equals 2% in period 𝑡. Assume that the aggregate production function is: 𝑌
𝑁 . The Central Bank follows a policy of adjusting the nominal interest rate to keep the real rate
fixed at a given value 𝑟. Suppose that there is a government stimulus program that increases public
spending from 𝑡 1 onwards.

A. Consider the IS‐LM graph. What is the impact of the increase in government spending on
GDP in 𝑡 1?

An increase in G pushes the IS curve to the right. The LM curve does not move since the Central
Bank didn’t change its policy. Thus, GDP in period 𝑡 1 increases, as seen in the graph below.

3
 B. What is the impact of the increase in government spending on unemployment in 𝑡 1?

Since 𝑌 𝑁 for output to increase employment must increase as well. Thus, the
unemployment rate will fall.

Consider the period 𝑡 2 equilibrium under the assumption that 𝜋 𝜋 and the Central Bank
adjusts the nominal policy rate to leave the real interest rate unchanged.

C. How does actual inflation in period 𝑡 1 compare to inflation in period 𝑡 , and how does
actual inflation in period 𝑡 2 compare to inflation in period 𝑡 1? To answer this question,
consider the Phillips curve.

We assumed that output started at its potential (“Start in the medium‐run equilibrium in year 𝑡”).
Also, note that we are told that actual and expected inflation equals 2% in period 𝑡. We also know
that the Phillips curve takes the form 𝜋 𝜋 𝑌 𝑌 . Hence, we can check that output is at its
potential in period 𝑡 using the Phillips curve: 2% 2% 𝑌 𝑌 ⇔𝑌 𝑌 .

Now use that 𝜋 𝜋 , so that the Phillips curve takes the form 𝜋 𝜋 𝑌 𝑌 .

We already mentioned that output increased from period 𝑡 to period 𝑡 1. As we can see from the
equation of the Phillips curve and the graph below, the new equilibrium point implies that 𝑌
𝑌 and so 𝜋 𝜋 0 ⇔ 𝜋 𝜋 .

In 𝑡 2, output is still above potential, as it was in 𝑡 1 (the equilibrium point in the graph below
is the same). Therefore, from the Phillips curve, the change in inflation is positive, i.e. the inflation

4
rate is higher in 𝑡 2 than it was in 𝑡 1.

D. Which should be the intervention of the Central Bank on the nominal interest rate to keep
the real interest rate constant?

From the Fisher relation, 𝑖 𝜋 𝑟 . When expected inflation is equal to last year’s inflation, the
Fisher relation becomes 𝑖 𝜋 𝑟 . So, as the inflation rate increases, to keep the real
interest rate constant, the Central Bank must increase the nominal interest rate.

E. Continue to period 𝑡 3. Making the same assumption about the way inflation expectations
are formed and about how the Central Bank conducts its policy (i.e. the Central Bank keeps
the real interest rate unchanged)

a. How does actual inflation in period 𝑡 3 compare to inflation in period 𝑡 2?

Inflation in 𝑡 3 is higher than inflation in 𝑡 2 since output is still above its natural level.

b. Suppose that in period 𝑡 4 the Central Bank increases the policy rate to restore
the original level of output. In the new equilibrium, how does inflation compare to
the original inflation level? What about consumption and investment? What are the
medium‐term effects of the fiscal expansion?

The increase in policy rate shifts the LM curve upwards. The economy moves along the Phillips curve
until the change in inflation is, again, zero, in the new equilibrium. However, the economy has
experienced an inflation increase for several periods. Therefore, the inflation rate will be higher in
the new equilibrium than in period 𝑡.

Consumption is a function of disposable income. Thus when the level of output returns to its original
level, so does consumption. Investment depends on output and on the interest rate. A higher
interest rate lowers investment. Therefore, investment will be lower than in the original equilibrium.

5
The medium‐term effect of the fiscal expansion are an increase of the real interest rate and
inflation rate, and reduction of investment (the decline of investment offsets the increase in
government expenditures such that output returns to its original level).

c. What should the Central Bank do if it wants to reach a new medium‐term equilibrium
with the same inflation (2%) as in the original equilibrium?

Initially, the Central Bank must further increase the interest rate. The 𝐿𝑀 curve shifts upwards to
𝐿𝑀′′. This creates a transitory bust (output is below potential). From the Phillips curve, this
decreases the inflation rate. The Central Bank can then shift the LM curve back to 𝐿𝑀′. If properly
done, the inflation rate is the same in this new equilibrium as it was in the original equilibrium.

6
Consider the same economy, subject to the same fiscal expansion, but now inflation expectations
are anchored at 2%. In particular, consider the period 𝑡 2 equilibrium under the assumption that
𝜋 2% (in this case, expected inflation is constant, as in the classical Phillips curve).

F. If the Central Bank leaves the real interest rate unchanged, how does (actual) inflation in
period 𝑡 1 compare to inflation in 𝑡? How does (actual) inflation in period 𝑡 2 compare
to inflation in 𝑡 1?

With inflation expectations anchored at 2%, the Phillips curve in any period 𝑡 is: 𝜋 0.02
𝛼 𝑢 𝑢 .

As before, the unemployment rate is lower in 𝑡 1 than in 𝑡 such that inflation is higher in 𝑡 1
than in 𝑡.

If the Central Bank keeps the policy rate, then the natural and actual unemployment rate will be the
same in 𝑡 1 and 𝑡 2. Then, from the Phillips curve, the inflation rate is the same in 𝑡 1 and
𝑡 2.

G. Which should be the intervention of the Central Bank on the nominal interest rate to keep
the real interest rate unchanged between 𝑡 1 and 𝑡 2?

From the Fisher relation with inflation expectations anchored at 2%, 𝑖 0.02 𝑟 . So, in this
case, to keep the real interest rate constant the Central Bank needs to keep the nominal interest
rate constant as well.

H. Continue to period 𝑡 3, but assume that the Central Bank has adjusted the policy rate to
restore output to its natural level. Making the same assumption about the level of expected
inflation and the real policy rate, how does actual inflation in period 𝑡 3 compare to the

7
 original inflation in period 𝑡?

Eventually, the central bank will need to increase 𝑖 (and thus 𝑟) for output to return to its natural
level. At that point, inflation is equal to its original level: with the classical Phillips curve as long as
there is no change in the unemployment/output gap there will be no change in inflation.

I. Compare your answers to H, Eb., and Ec. above. Discuss the importance anchoring inflation
expectations?

With anchored inflation expectations, it is easier to stabilize the level of inflation. In this example, if
inflation expectations are not anchored it is harder for the central bank to return to the original level
of inflation (i.e., it needs to engineer a transitory recession in order to drag inflation expectations
back to 2%, generating more volatility in output). There is no need to do so if inflation expectations
are anchored.

IS2

LM2
IS1
LM1

0
Yn Y

π – πe PC

0 Y
Yn

8
Universitat Pompeu Fabra 2022-23

SOLUTIONS PROBLEM SET n. 8

Question 1

In answering this question, please make the following three assumptions:

a. Agents expect inflation in year 𝑡 1 to be equal to actual inflation in year 𝑡,
𝜋 𝜋 𝜋
b. The aggregate production function is the following: 𝑌 𝐴𝑁, where 𝑌 is GDP, 𝑁 is the number of
workers, and 𝐴 is productivity per worker.
c. The labour force is constant.

1. Show in a graph the IS curve, the LM curve and the Phillips curve. Show the equilibrium of the economy
under the assumption that GDP is at its natural level and inflation is stable.

2. Assume that GDP in Jan. 2021 in the US was at its natural level and inflation was stable. With the
introduction of the American Rescue Plan Act of 2021, President Joe Biden publicly unveiled his COVID-19
stimulus proposal, which includes $1.9 trillion in funding. The plan pledges to provide various benefits, such
as $1,400 stimulus payments, extended unemployment benefits, a vaccine distribution effort, and other
measures. Show how this policy affects the equilibrium in the IS-LM-PC model. What happens to
consumption, GDP, investment, unemployment and inflation in the period immediately after the
implementation of the policy (i.e., period t+1) if the Central Bank does not react?

1
 IS’

A’

A’

The IS curve will shift upward. GDP, investment and consumption will increase, unemployment will go down,
and from PC curve inflation will start increasing. 𝐴 will represent the new equilibrium if there is no response
from central banks.

3. Now assume that the Central Bank does not want inflation to deviate from its targeted level, 2%. How will
the Central Bank react to the policy of Joe Biden? Show the new medium-term equilibrium using the IS-LM-
PC model.

IS’

A’’
LM ’
A’

A’

The Central Bank will react with a contractionary policy. The LM curve will shift upward. At the new
equilibrium at 𝑨 , GDP will stay at its natural level and inflation will be stabilized at the new natural level of
real interest rate.

4. Suggest an alternative policy that Joe Biden could have implemented to reduce the unemployment rate in
the US without inducing the Central Bank to implement a contractionary monetary policy.

Any policy that affects the natural rate of unemployment and potential output. Examples can be
investments into research and development (which increases labour productivity 𝐴), liberalization in
different product markets or reductions in import tariffs (which reduces mark-ups 𝜇), or reduction in
minimum wages or employment protection (which reduces 𝑧). All these policies imply an increase in
potential output, thus shifting the PC outward. Inflation will move downward inducing the Central

2
Bank to reduce interest rates. Eventually, the economy will move to a new equilibrium in which GDP
is higher and inflation is stable. Also, the unemployment rate will be lower.

Question 2
Assume that: the aggregate production function takes the form 𝑌 𝐹 𝐾 ,𝑁 𝐾 𝑁 with 0
𝛼 1; capital depreciates at a rate 𝛿; there is no government spending or taxes; the labor force is
constant, i.e. 𝑁 𝑁; aggregate saving is a constant fraction 𝑠 (the saving rate) of aggregate output.
Denote with 𝑘 ≡ 𝐾 /𝑁 the amount of capital per worker and with 𝑦 ≡ 𝑌 /𝑁 output per worker.

a) What is the marginal product of capital (MPK)? Is it constant or decreasing in 𝐾 ? Write the
production function in per capita (per worker) terms, i.e. 𝑦 as a function of 𝑘 .

𝜕𝑌 𝐾
𝑀𝑃𝐾 ≡ 𝛼𝐾 𝑁 𝛼
𝜕𝐾 𝑁
𝜕𝑀𝑃𝐾 𝐾
1 𝛼 𝛼 0
𝜕𝐾 𝑁
Thus, MPK is decreasing in 𝐾 .

𝑌 𝐾 𝑁 𝐾
𝑘 ⇒ 𝑦 𝑘
𝑁 𝑁 𝑁

b) Obtain the equation that determines the change in capital per worker 𝑘 – 𝑘 and, from
that, the equation that determines the growth rate of capital per worker 𝑘 – 𝑘 /𝑘 . Then
∗
show the graphical representation of each, indicating the steady state level 𝑘 .

𝐾 1 𝛿 𝐾 𝐼 ⇒ 𝐾 𝐾 𝐼 𝛿𝐾
As
𝐼 𝑠𝑌 ⇒ 𝐾 𝐾 𝑠𝑌 𝛿𝐾

Divide by N and use 𝑦 𝑘 :

𝑘 𝑘 𝑠𝑘 𝛿𝑘

The absolute change in 𝑘 is the difference between the investment curve and the depreciation line.

3
Divide by 𝑘 :
𝑘 𝑘 𝑠
𝛿
𝑘 𝑘

The growth rate of 𝑘 is the difference between the investment/capital ratio curve and the depreciation
rate.

c) How is the growth rate of output per worker 𝑦 – 𝑦 /𝑦 related to the growth rate of
capital per worker 𝑘 – 𝑘 /𝑘 ?

Denote with 𝑔 the growth rate of output per worker, with 𝑔 the growth rate of capital per worker,
so that:
𝑦 1 𝑔 𝑦
And
𝑘 1 𝑔 𝑘

Using the equality we get, 𝑦 𝑘 , for period 𝑡 1:

𝑦 𝑘

Then
𝑦 𝑘 𝑦 𝑘
⇒ 1 1 1 1
𝑦 𝑘 𝑦 𝑘
⇒ 1 𝑔 1 𝑔

Taking the logs in both sides:

log 1 𝑔 𝛼 log 1 𝑔 ⇒ 𝑔 𝛼𝑔

4
 d) Solve for the steady state level of capital per worker 𝑘 ∗ and output per worker 𝑦 ∗ , as a function
of the parameters of the model (𝛼, 𝛿, 𝑠). Then compute the value of 𝑘 ∗ and 𝑦 ∗ if, for example,
𝛼 1/2, 𝛿 1%, and 𝑠 0.10.

Steady state: 𝑘 𝑘 . From 𝑘 𝑘 𝑠𝑘

𝛿𝑘 this implies:
𝑠
0 𝑠𝑘 𝛿𝑘 ⇒ 𝑘
𝛿
𝑠 /
⇒ 𝑘∗
𝛿
And
𝑠 /
𝑦∗ 𝑘∗
𝛿

For the given parameters:

𝑠 0.1 .
𝑘∗ 100
𝛿 0.01
.

∗ ∗
0.1 .
𝑦 𝑘 10
0.01

e) Assume the economy is at the steady state when, in period 𝑇, a permanent increase in the
saving rate 𝑠 occurs (i.e. the new saving rate is 𝑠 𝑠). Show analytically (with your results
from (d)) and graphically (with the graphs from (b)) how this change affects the long run
levels of capital and output per worker (the new steady state) and the growth rate during the
transition.

Consequences of higher saving rate:

 higher long run levels of capital and output:

/
∗∗
𝑠 𝑠 /
𝑘 𝑘∗
𝛿 𝛿
/
∗∗
𝑠 𝑠 /
𝑦 𝑦∗
𝛿 𝛿
 higher (positive) growth rates during transition to the new steady state:

𝑘 𝑘 𝑠 𝑠
𝑔 𝛿 𝛿 𝑔
𝑘 𝑘 𝑘
And
𝑔 𝛼𝑔 𝛼𝑔 𝑔

though eventually the growth rate will go to zero again.

5
 f) If for example the new saving rate is 𝑠 0.50, what will be the new steady state levels 𝑘 ∗∗
and 𝑦 ∗∗ towards which the economy will converge? Compute also the growth rate of 𝑘 when
the change in 𝑠 occurs (at time 𝑇), starting from the steady state 𝑘 ∗ , and immediately after (at
time 𝑇 1 and 𝑇 2): what is the response of the growth rate to the change?

0.5 .
𝑘 ∗∗ 2500
0.01
And
. / .
0.5
𝑦 ∗∗ 50
0.01

Growth rate at time 𝑇, when 𝑘 𝑘∗ 100:

𝑠 0.5
𝑔 𝛿 .
0.01 4%
𝑘 100

Growth rate at time 𝑇 1, when 𝑘 100 1 0.04 104:

𝑠 0.5
𝑔 𝛿 0.01 3.9%
𝑘 104 .

Growth rate at time 𝑇 2, when 𝑘 104 1 0.039 108.1:

𝑠 0.5
𝑔 𝛿 0.01 3.8%
𝑘 108.1 .

i.e. growth rate jumps up becoming positive when s increases, then slowly decreases towards zero as
we approach the new higher steady state.

6
Question 3
Consider two countries (A and B) that are identical in all respects, in particular they both have the
same economic structure described at the beginning of the previous exercise, except that country A
is initially richer than country B, i.e. it has higher initial capital and output per worker: 𝑘
𝑘 ,𝑦 𝑦 .
a) Compare the growth rates of the two countries: will they grow at the same rate during the
transition? will one grow faster than the other?

Growth rate it is inversely related to the level of k (because of decreasing returns to capital), so country
B with lower capital per worker has a higher growth rate than country A during the transition:
𝑠 𝑠
𝑔 𝛼𝑔 𝛿 𝛿 𝛼𝑔 𝑔
𝑘 𝑘

b) What will happen in the long run? Will the poorer country catch up or will it remain poorer?
Explain.

In the long run, both countries reach the same steady state level of capital per worker 𝑘 ∗ and therefore
the same output per worker 𝑦 ∗ (since they have the same technology, the same , , 𝑠). By growing
faster, the initially poorer country catches up (i.e. there is convergence across these countries).

c) There is evidence of convergence among rich countries (OECD) and also some emerging
economies; but many of the poorest economies (in Africa and elsewhere) have not been converging.
How do you explain this apparent contradiction with your results in a) and b)?

The model predicts conditional convergence, i.e. growth rates are inversely related to the level of
capital per worker conditional on all the relevant structural and policy parameters that determine the
steady state. But if countries differ in their policies, saving rates, technology, and other factors that
affect productivity, they will not be converging to the same steady state (or to the same balanced
growth path). It must be that poorer countries that are not growing fast are not only behind in their
historical starting level, but they also have less efficient technologies and institutions, lower human
capital, lower saving rates, etc..

d) The World Bank has the important task of facilitating the development of poor countries, but there
is often debate over what kinds of policies can effectively achieve this goal. One policy option is to
provide donations or aid to poor countries, but the long-run effectiveness of such policies is not
always clear. Please discuss: what is the long-term effect of a World Bank donation in this model?

As we discussed previously in the standard Slow Model framework, people typically have a fixed

7
saving rate, 𝑠, regardless of their income level, and given a specific depreciation rate, 𝛿, a country
will eventually converge to a steady-state equilibrium. Consequently, policies such as donations or
aid to poor countries, which temporarily increase their capital stock (e.g. from 𝑘 to 𝑘 ), will not
have a sustained effect on the country's long-term growth.

Question 4

Assume that: the aggregate production function takes the form 𝑌 𝐹 𝐾 , 𝑁 =20𝐾 𝑁 , with capital
depreciates at a rate =0.1; there is no government spending or taxes; the labor force is constant, i.e.
𝑁 𝑁. Finally, aggregate saving is a constant fraction s=0,1 of aggregate output if 𝑌/𝑁 200 ;
if 𝑌/𝑁 200, however, aggregate savings is equal to zero. You can interpret this as a situation in
which the country cannot save if income is below some “subsistence” level. Denote with kt the amount
of capital per worker (Kt /N) and with 𝑦 output per worker (𝑌 /𝑁).

a) Obtain the equation that determines the change in capital per worker (kt+1 – kt) and, from that, the
equation that determines the growth rate of capital per worker (kt+1 – kt) / kt. Then show the graphical
representation of each, indicating the steady state level 𝑘 ∗ . Explain the dynamic evolution of capital
and output depending on initial conditions.

We have two cases:

1. In case of 𝑌/𝑁 200, we have 𝑠 0.1, thus:

𝐾 1 𝛿 𝐾 𝐼 ⇒ 𝐾 𝐾 𝐼 𝛿𝐾

𝐼 𝑠𝑌 ⇒ 𝐾 𝐾 𝑠𝑌 𝛿𝐾 0.1𝑌 0.1𝐾

.
Divide by N and use 𝑦 20𝑘 , we obtain:
.
𝑘 𝑘 2𝑘 0.1𝑘

From this dividing by 𝑘 yields:

𝑘 𝑘 2
0.1
𝑘 𝑘 .

2. While in case of 𝑌/𝑁 200, we have s=0, thus:

𝐾 1 𝛿 𝐾 𝐼 ⇒ 𝐾 𝐾 𝐼 𝛿𝐾

𝐼 𝑠𝑌 ⇒ 𝐾 𝐾 𝑠𝑌 𝛿𝐾 0.1𝐾
.
Divide by N and use 𝑦 20𝑘 , we obtain:
𝑘 𝑘 0.1𝑘

From this dividing by 𝑘 yields:

𝑘 𝑘
0.1
𝑘

Thus, we have a break in the savings function at

8
 .
𝑦 20𝑘 200 ⇒ 𝑘 100

As visible from the graphical representation, this implies that there are two steady states of the
economy:

0  0.1k *  k *  0  y *  0
2
0  **0.5 0.1  k **  400  y **  20k **0.5  400
k
That is, when the economy starts below the subsistence level, the country cannot save, and due to
depreciation the capital stock converges to zero. In contrast when the economy starts above the
subsistence level, the economy will grow each year and will converge gradually to the steady state.

b) Suppose that this country starts out with a relatively low capital stock, of kt =81. The World Bank
wants to “donate” resources to increase the country’s initial capital stock to kt =100. What would be
the long-term consequences of this donation for the economy?

A donation that would push the country’s stock of capital per worker to the level that produces output
above the subsistence level would put the country on a growth path that converges eventually to a
steady state of:

9
 𝑦 ∗∗ 400
Without the support from the World Bank, the country’s economy would remain below the
subsistence level, and the economy would gradually converge to:

𝑦∗ 0

10