• Changes in the demand for goods lead to changes in production.
• Changes in production lead to changes in income.
• And changes in income lead to changes in the demand for goods.
Nothing makes the point better than this cartoon.
3.1 THE COMPOSITION OF GDP
The purchase of a machine by a firm, the decision to go to a restaurant by a consumer, the purchase ol combat aeroplanes by the federal government—these are clearly very different decisions and depend on different factors. So, il we want to understand what determines the demand lor goods, it makes sense to decompose aggregate output (GDP) Irom the point of view of the different goods being produced, and from the point ot view of the different buyers lor these goods.
The decomposition of GDP typically used by macroeconomists is given in Table 3.1. (A more detailed version, with more formal definitions, is given in Appendix I at the end of the book.)
• first comes consumption (which we will denote by the letter С when we use algebra throughout this book). These arc the goods and services purchased by consumers, ranging Irom lood to airline tickets, to holidays, to new cars, and so on. Consumption is by far the largest component of GDP. In 2008, it accounted lor 50 per cent ol GDP in Ausiralia.
• Second comes investment 1/), sometimes called fixed investment land sometimes called gross fixed capital formation') to distinguish il from inventory investment (which we will discuss shortly). Investment is the sum of non-residential investment, the purchase by lirms ot new factories or new machines Irom turbines to computers), and residential investment, the purchase by people ol new houses or apartments.
Non-residential investment and residential investment, and ihe decisions behind them, have more in common than might lirsi appear. Firms buy machines or factories to be able to produce output in the future. People buy houses or apartments lo gel housing services in the future. In both cases, the decision to buy depends on the services these goods will yield in the future,- so it makes sense to treat them together. Together, non-residential and residential investment accounted for 22 per cent of GDP in 2008.
• Third comes government spending G). This represents the purchases of goods and services by the federal, state and local governments. The tederal or Commonwealth government spends more than a third ol G. The goods range from aeroplanes to office equipment. The services include services provided by government employees. In effect, the national income accounts treat the government
Table 3.1 The composition of Australian GDP. 2008
Chain volume measures $ billion % of GDP
GDP 1,129 100%
1. Consumption 627 56%
2. Investment 245 22%
Non-residential 177 16%
Residential 68 6%
3. Government spending 250 22%
4. Net exports -18 -2%
Exports 235 21%
5. Inventory investment Imports 4 -253 0% -22%
1 1
SOURCE RBA Bul/elm.Table G11. year to June 2008.
as buying the services provided by government employees—and then providing these services to the public, Iree of charge.
Note that G doesn't include government transfers such as social security and wellare payments nor interest payments on the government debt. Although these are clearly government expenditures, they aren't purchases ol goods and services. To include them would effectively mean they would be double-counted in consumption. That is why the number lor government spending on goods and services in Table 3.1, 22 per cent ol GDP in 2008, is smaller than the total lor government spending including transfers and interest payments. That number, in 2008, was 32 per cent of GDP.
• The sum ol lines 1, 2 and 3 gives the purchases of goods and services by Australian consumers, lirms and the government. To get to the purchases of Australian goods and services, two more steps arc- needed:
First we must subtract imports /ЛI ■ the purchases ol foreign goods and services by Australian- resident consumers, lirms and the government.
Second, we must add exports 'X the purchases ol Australian goods and services by foreigners.
The diflerence between exports and imports, X - Ш), is called net exports, or the trade balance. If exports exceed imports, a country is said to run a trade surplus. II exports are less than imports, the country is said to run a trade deficit. In 2008, Australian exports accounted for just below 21 per cent ol GDP. Australian imports were just above 22 per cent ol GDP, so Ausiralia was mnning a trade deficit almost equal to 2 percent ol GDP.
• So far we have looked at various sources ol purchases (equivalentlv, sales / ol Australian goods and services in 2008. To get to Australian production in 2008. we need one last step.
In any given year, production and sales needn t be equal. Some ol the goods produced in a given year aren't sold in that year, but are sold in later years. And some of the goods sold in a given year may have been produced in an earlier year. The dillerence between goods produced and goods sold in a given year—equivalentlv, the difference between production and sales—is called inventory investment. Il production exceeds sales firms accumulate inventories: inventory investment is positive. If production is less than sales, firms decrease inventories: inventory investment is negative.
Inventory investment is typically small—positive in some years, negative in others. In 2008, inventory investment was positive, and equal lo close to 0 per cent of GDP. Put another way, prod.iction was higher than sales by an amount equal to just above 0 per cent ol GDP.
We now have what we need lo develop our first model ol output determination.
3.2 THE DEMAND FOR GOODS
Denote the total demand for goods by Z. Using the decomposition of GDP we saw in Section 3.1. we
can write Z as
ZsC + f + G + X- IM
This equation is an identity which is why il is written using the symbol '= rather than an equal sign).
It defines Z as the sum of consumption, plus investment, plus government spending, plus exports minus
imports.
We now need ю think about the determinants of Z. To simplify our task lets first make a number
of simplifications:
• Assume thai all firms produce the same good which can be used by consumers for consumption, by lirms for investment or by the government. With this big simplification, we need to look at only one market—the market for the good—and think about what determines supply and demand in that market.
Exports - imports = net ► exports £ trade balance Exports > Imports о trade surplus Exports < Imports » trade deficit
Make sure you ► understand each of these three equivalent ways of stating the relation between production, sales and inventory investment; Production - sales = inventory investment Production = sales + inventory investment Inventory investment = production - sales
A model nearly always starts with the word ^ Assume (or Suppose).
This s an indication that reality is about to be simplified in order to focus on the issue at hand.
• Assume that lirms are willing to supply any amount of the good at a given price, P. This assumption allows lis to focus on the role of demand in the determination of output. As we will see later in the
book, this assumption is valid only in the short run. When we move to study how the economy performs in getting to the medium run (starting in Chapter 6), we will need to give it up. But, for the moment, the assumption will simplify our life.
• Assume that the economy is closed—that is, it doesnt trade with the rest of the world. Both exports and imports are zero. This assumption clearly goes against the facts, as modern economies trade with the rest ot the world. I.ater on 'starting in Chapter 18), we will abandon this assumption and look at what happens when the economy is open. But, tor the moment, this assumption will also simplify our life—we won't have to think about what determines exports and imports.
Under the assumption that the economy is closed, X - /М - 0, so the demand for goods Z is simply the sum of consumption, investment and government spending:
Z = С + 1 + G
Let's now discuss each ol these three components in turn. Consumption (C)
Consumption decisions depend on many factors. But the main one is surely income, or, more precisely, disposable income, the income that remains once consumers have received transfers from the government and paid their :axes. When their disposable income goes up, people buy more goods,- when it goes down, they buy fewer goods.
Let С denote consumption and Yp denote disposable income. We can then write
С = C(Vp) (3.1)
(+)
1 his is a formal way of slating that consumption. C, is a function of disposable income. Yp. The lunction C(Yp) is called the consumption function. The positive sign below Yp reflects the tact that when disposable income increases so does consumption. Economists call such an equation a behavioural equation to indicate lhai the equation captures some aspect of behaviour—in this case, the behaviour ol consumers.
We will use functions in this book as a way of representing relations between variables. What you need to know about functions—which is very little —is described in Appendix 2 at the end of this book. This appendix develops the mathematics you need to use this book. Dont worry—we will always describe a function in words when we introduce it lor the lirst lime.
It is often useful to be more specific about the form of the function. Here is such a case. It is reasonable to assume that the relation between consumption and disposable income is given by
С = c0 + r,Yp (3.2)
In words: It is reasonable to assume that the function is a linear relation. The relation between consumption and disposable income is then characterised by two parameters, c0 and r,:
• The parameter c, is called the propensity to consume. It is also called the marginal propensity to consume. We will drop marginal for simplicity.) It gives the effect of an additional dollar of disposable income on consumption. И c, is equal to 0.6, then an additional dollar ol disposable income increases consumption by $0.6 60 cents.
A natural restriction on c', is that it be positive—an increase in disposable income is likely to lead to an increase in consumption. Another natural restriction is that c, be less than 1—people are likely to consume only part of any increase in disposable income, and to save the rest.
• The parameter Гц has a simple interpretation. It is what people would consume it their disposable income in the current year were equal to zero: il Yp equals zero in equation (.3.2), С = f(1.
©
It is quite possible, il current income is equal to zero, that consumption is still positive: people still need to eat! This would imply that r„ is positive. How can people have positive consumption if their
income is equal to zero? Answer: They dissave. They consume either by selling some of their assets or by borrowing. II cn is negative, it implies that people don't expect in aggregate to be able to sell asse:s or to borrow if aggregate disposable income became zero.
The relation between consumption and disposable income implied by equation (3.2.1 is drawn in Figure 3.1. Because it is a linear relation, it is represented by a straight line. Its intercept with the vertical axis is c,,; its slope is c,. Because r, is less than I, the slope ol the line is less than 1. Equivalently, the line is flatter than a 45-degree line. iA relresher on graphs, slopes and intercepts is also given in Appendix 2.)
Next we need to define disposable income, Yp. Disposable income is given by
yD S V - T
where У is income and Tis taxes paid minus government transfers received by consumers. For short, we will refer to T simply as taxes—but remember that it is equal to taxes minus transfers. Note that the equation is an identity, indicated by the symbol
(3.3)
Replacing Yp in equation : 3.2) gives
С = с, + c,(Y-T)
liquation (3.3) tells us that consumption, C, is a function of income, Y, and taxes, T. Higher income increases consumption, although less than one-for-one. Higher taxes decrease consumption, also less than one-for-one.
In Australia, the main t taxes paid are income taxes ($120 billion in 2007), company taxes ($60 billion) and indirect taxes such as the GST ($1 IS billion).The main sources of government transfers are social security and unemployment benefits. In 2007, transfers to individuals were $100 billion (25% of which was for the aged pension, and 5% for unemployment benefits).
Investment (/)
(3.4)
Models have two types of variables. Some variables depend on other variables in the model, and are therefore explained within the model. Such variables are called endogenous. This was the case lor consumption above. Other variables are not explained within the model but are instead taken as given. Such variables are called exogenous. This is how we will treat investment here. We will take investment as given, and write
1 I
Figure 3.1
Consumption
and disposable
income
U
с
л
•о
а. ^^^ Consumption
Е ^^^^ function
3
и С
О
и slope =с1
Со
Disposable income, YD
Consumption increases with disposable income, but less than one-for-one.
Putting a bar over investment is a simple typographical way to remind us that we take investment as 4 Endogenous variables: given.
taxes minus government transfers.
We take investment as given to keep our model simple. But the assumption isn t innocuous. It implies that, when we look later at the ellects ol changes in production, we will be assuming that investment doesn't respond to changes in production. It isn't hard to see that this implication may be a bad description of reality. I inns that experience an increase in production may decide that they need more machines, and increase their investment We leave this mechanism out of the model for the moment,- we will introduce a more realistic treatment of investment in Chapter 5.
Government spending (G)
The third component of demand in our model is government spending, G. Together with taxes, T. G 4 Recall:Taxes stand for describes fiscal policy—the choice of taxes and spending by the government. Just as we did for investment, we will take С and T as exogenous. But the reason we assume G and T are exogenous is diflerent from the reason we assumed investment is exogenous. It is based on two arguments:
• First, governments dont behave with the same regularity as consumers or lirms, so there is no reliable rule we could write lor G or T corresponding to the rule we wrote, for example, for consumption. (The argument isn't hilly convincing. Even if governments don't follow simple behavioural rules as consumers do, a good part of their behaviour is predictable. We will look at these issues later, in particular in Chapters 25 to 27, but we leave them aside until then.)
4 Because we will (nearly always) take G and T as exogenous, we won't use a bar to denote their value. This will keep the notation lighter.
Think of an economy that produces only haircuts.There cannot be inventories of haircuts—how can there be haircuts produced but not sold? Equilibrium requires producdon of haircuts to be equal to demand for haircuts. We will come back later to what happens when firms can hold inventories ard so production needn't be 4 equal to demand.
• Second, and more importantly, one ol the tasks of macroeconomists is to think about the implications ol alternative spending and tax decisions. We want to be able to say, II the government were lo choose these values lor G and 7", this is what would happen . The approach in this book will typically treat G and T as variables chosen by the government, and not try to explain them within the model.
3.3 THE DETERMINATION OF EQUILIBRIUM OUTPUT
Let's collect and put together the pieces we have introduced so lar.
those explained within the model. Exogenous variables: those taken as given.
Assuming thai exports and imports are both zero, the demand for goods is the sum of consumption, investment and government spending:
Z s С + I + G
Replacing С and I Irom equations (3.31 and (3.4), we get
Z = c„ + c,(Y - T) + 1 + G
(3.5)
The demand for goods, Z. depends on income, V, taxes, 7", investment, /, and government spending, G.
(3.6)
Tnree types of equations:
• identities
• behavioural equations 4 • equilibrium conditions.
Let's now turn to equilibrium in the goods market, and the relation between production and demand. II firms hold inventories, then production needn't be equal to demand. For example, firms can respond to an increase in demand by drawing down inventories, by having negative inventory investment. They can respond to a decrease in demand by continuing to produce and accumulate inventories, by having positive inventory investment. It will be helpful to ignore this complication here, and to start by assuming that firms dont hold inventories. In this case, inventory investment is always equal to zero, and equilibrium in the goods market requires that production, Y, be equal to the demand for goods, Z:
V = z
I his equation is called an equilibrium condition. Models include three types of equations— identities, behavioural equations and equilibrium conditions. You now have seen examples ol each. The equation defining disposable income is an identity,- the consumption function is a behavioural equation,- and the condition that production equals demand is an equilibrium condition.
©
Replacing demand Z in equation (3.6) by its expression from equation (3.5) gives
У = c0 + c,(Y-T) + 7 * G (3.7)
Equation <3.7) represents algebraically what we stated informally at the beginning of this chapter:
Relate this statement to ► 111 equilibrium, production, У Ithe left side of the equation), is equal to demand lllie right sideI Demand the cartoon at the start in turn depends on income, У, which is itself equal to production.
Note that we are using the same symbol, У, tor production and income. This is no accident! As you saw in Chapter 2, we can look at GDP from either the production side or the income side. Production and income are identically equal.
Having constructed a model, we can solve it to look at what determines the level of output—how output changes in response to, say, a change in government spending. Solving a model means not only solving it algebraically but also understanding why the results are what they arc. In this book, solving a model will also mean characterising the results using graphs—sometimes skipping the algebra altogether—and describing the results and the mechanisms in words. Macroeconomists always use these three tools:
1. Algebra to make sure that the logic is correct.
2. Graphs to build the intuition.
3. Words to explain the results. Make it a habit to do the same.
Using algebra
Rewrite the equilibrium equation (3.7):
У = c() + с,(У-Л + I
(I
Move С|У to the lelt side and reorganise the right side:
Y-Co + T+ G- c,T
Divide both sides by (I - r,):
[Со + 1 + Ci - с J]
- С,
Equation (3.8) characterises equilibrium output, the level of output such that production equals demand. Let's look at both terms on the right, beginning with the second term.
• The term [
• Turn to the first term, l/( I - c,). Because the propensity to consume (c,) is between zero and one, !/(1 — r,) is a number greater than one. For this reason, this number, which multiplies autonomous spending, is called the multiplier. The closer c, is to one, the larger the multiplier.
У =
(3.8)
Autonomous means independent—in this case, independent of output.
If Г = G. then (G-c,T) = (G-7) + (I -Ci)r = (I - C|)T > 0. ►
The distance between Y> and V" on the vertical axis is larger than the distance between A and В—which is equal to I billion.
©
What docs the multiplier imply? Suppose that, lor a given level of income, consumers decide to consume more. More precisely, assume that c„ in equation (3.3) increases by $1 billion. Equation (3.8) tells us that output will increase by more than $1 billion. For example, il c( equals 0.5, the multiplier equals |/( I - 0.5) : 2. so that output increases by 2 X 1 billion = $2 billion.
We have looked at an increase in consumption, but equation (3.8) makes clear that any change in autonomous spending—from a change in investment, to a change in government spending, to a change in taxes—will have the same qualitative effect: it will change output by more than its direct effect on autonomous spending.
Where does the multiplier effect come from? Looking back at equation (3.7) gives the clue—an increase in r,p increases demand. The increase in demand then leads to an increase in production and income. But the increase in income further increases consumption, which further increases demand, and so on. Lhe best way to strengthen this intuition is lo represent the equilibrium using a graph. Let's now do that.
Using a graph
Let's characterise the equilibrium graphically.
• First, plot production as a lunction ol income. In Figure 3.2, measure production on lhe vertical axis. Measure income on the horizontal axis. Plotting production as a function of income is straightforward. Recall that production and income are always equal. Thus, the relation between them is the 45-dcgrce line, the line with a slope equal to 1.
(3.9)
Figure 3.2 Equilibrium in the goods market
У
Income, Y
Equilibrium output is determined by the condition that production be equal to demand.
• Second, plot demand as a function of income. The relation between demand and income is given by equation (3.5 . Lets rewrite it here tor convenience, regrouping the terms for autonomous spending together in the term in parentheses:
Z - (c0 + I + G-cJ) + c,Y
Demand depends on autonomous spending, and on income—through its effect on consumption. The relation between demand and income is drawn as ZZ in the graph. The intercept with the vertical axis—the value of demand when income is equal to zero—equals autonomous spending. The slope of the line is the propensity to consume, c,: when income increases by I demand increases by C|. Under the restriction that i", is positive but less than I, the line is upward sloping but with slope less than 1.
• In equilibrium, production equals demand. Thus, equilibrium output, Y, is given by the intersection ol the 45-degree line and the demand relation, at point A. To the left ot A, demand exceeds production,- to the right of A, production exceeds demand. Only at A are demand and production equal.
Now, suppose that c0 increases by $1 billion. At the initial level ol income the level ot income associated with point A), consumers increase their consumption by $1 billion. What happens then is shown in Figure 3.3, which builds on f igure 3.2.
Equation (3.9) tells us that lor any value of income, demand is higher by $1 billion. Before the increase in c0, the relation between demand and income was given by the line ZZ. After the increase in f„ by $1 billion, the relation between demand and income is given by the line ZZ', which is parallel to ZZ but higher by $1 billion. In other words, the demand relation shifts up by $1 billion. The new equilibrium is at the intersection of the 45-degree line and the new demand relation, at point A'.
Equilibrium output increases from Y to Y'. I he increase in output, V' - Y), which we can measure on either the horizontal or the vertical axis, is larger than the initial increase in consumption of $1 billion. This is the multiplier effect.
With the help of the graph, it becomes easier to tell how and why the economy moves Irom A to A'. The initial increase in consumption leads to an increase in demand of Si billion. At the initial level ot income, Y, the level ol demand is now given by point В—demand is $1 billion higher. To satisfy this higher level ol demand, firms increase production by $1 billion. The economy moves to point C, with both demand and production higher by $1 billion. But this isn't the end ol the story. The higher level of production implies an increase in income ol $1 billion 'recall that income = production), and to a lurthcr increase in demand, so demand is now given by point D. Point D leads to a higher level ol production, and so on, until the economy is at A' where production and demand are again equal, and is therefore the new equilibrium.
We can pursue this line of explanation a bit further, and this will give us another way ot thinking about the multiplier.
Trick question:Think ► about the multiplier as the result of these successive rounds. What would happen in each successive round if c,.
the propensity to consume, were larger than one!
• The first-round increase in demand given by the distance AB in Figure 3.3, equals $1 billion.
Income, У
Figure 3.3 The effects of an increase in autonomous spending on output
©
An increase in autonomous spending has a more than one-for-one effect on equilibrium output
• This first-round increase in demand leads to an equal increase in production, also given by the distance AR, thus $ I billion.
• This first-round increase in production leads to an equal increase in income given by the distance ВС, also equal to $1 billion.
• The second-round increase in demand, given by the distance CD. equals $1 billion (the increase in income in the lirst round) times the propensity to consume out ol income, c, —hence, $c, billion.
• This second-round increase in demand leads to an equal increase in production, also given by the distance CD, and thus an equal increase in income, given by the distance DE.
• The third-round increase in demand equals $f, billion 1 the increase in income in the second round), times c,, the marginal propensity to consume out ol income,- it is equal to $i', x с, ^ $r,2 billion, and so on.
Following this logic, the total increase in production alter, say, it rounds equals $1 billion times the sunt:
1 + c, + c,2 + ... + с,"-1
Such a sum is called a geometric series. Geometric series will come up often in this book. iA refresher is given in Appendix 2. i One main property is that, when f, is less than one (as it is here) and as ii gets larger and larger, the sum keeps increasing but approaches a limit. That limit is l/( I — с,), making the eventual increase in output equal to $1/(1 — t",1 billion.
I he expression 1/(1 — fj) should be familiar-—it is the multiplier derived another way. This gives us an equivalent, but more intuitive, way of thinking about the multiplier. We can think of the original increase in demand as triggering successive increases in production, with each increase in production implying an increase in income, which leads to an increase in demand which leads to a hirther increase in production, which leads. . . and so on. The multiplier is the sum of all these successive increases in i In the model we saw production. earlier, we ruled out this
possibility by assuming
Using words firms didn>t ho,d
inventories and so
How can we summarise our findings in words? couldn't rely on
Production depends on demand, which depends on income, which is itsell equal to production. An inventories to respond increase in demand, such as an increase in government spending, leads to an increase in production and t0 demand a corresponding increase in income. This increase in income leads to a further increase in demand, which leads to a further increase in production, and so on. The end result is an increase in output that is larger than the initial shift in demand by a lactor equal to the multiplier.
The size of the multiplier is directly related to the value of the propensity to consume—the higher the propensity to consume, the higher the multiplier What is the value of the propensity to consume in Australia today? To answer this question, and more generally to estimate behavioural equations and their parameters, economists use econometrics, the set of statistical methods used in economics. To give you a sense ol what econometrics is and how it is used, read Appendix 3. This appendix gives you a quick introduction, using as an application the estimation of the propensity to consume. The conclusion from a standard estimation in the appendix is that, in Australia today, the propensity to consume is around 0.55. An additional dollar ol income leads on average to an increase in consumption ol 55 cents. This implies a multiplier equal to 1/(1 - C,) = 1/(1 - 0.55) - 2.22
How long does it take for output to adjust?
Let's return to our example one last time. Suppose that t'(, increases by $1 billion. We know that output will increase by an amount equal to the multiplier 14 - t-, limes $1 billion. But how long will it take for output to reach this higher value?
Llnder the assumptions we have made so tar, the answer is: Right away! In writing the equilibrium condition (3.6), we have assumed that production is always equal to demand—in other words, production responds to demand instantaneously. In writing the consumption lunction (3.21, we have assumed that consumption responds to disposable income instantaneously. Llnder these two
assumptions, the economy goes instantaneously from point A to point A' in Figure 3.3. The increase in demand leads to an immediate increase in production the increase in income associated with the increase in production leads to an immediate increase in demand, and so on. We can think, of the adjustment in terms of successive rounds as we did earlier, but all these rounds happen at once.
This instantaneous adjustment doesn't seem plausible. And it isn't. A firm that faces an increase in demand may decide to wait before adjusting its production, meanwhile drawing down its inventories to satisfy demand. A worker who gets a pay raise may not adjust her consumption right away. And these delays imply that the adjustment of output will take time.
Describing formally this adjustment ot output over time—describing what economists call the dynamics of adjustment—would be loo hard. But it s easy to do it in words:
• Suppose, for example, that firms make decisions about their production level at the beginning ol each quarter,- once the decision is made production cannot be adjusted for the rest ol the quarter. Il purchases are higher than production, firms draw down inventories to satisfy purchases. Il purchases are lower than production, firms accumulate inventories.
• Now, return to our example, and suppose consumers decide to spend more,- thev increase t"(,. During the quarter in which this happens, demand increases but—because of our assumption that production was set at lhe beginning of the quarter—production doesn't yet change. Therefore income doesn't change either.
• In the following quarter, lirms having observed an increase in demand in the previous quarter are likely to set a higher level ot production. This increase in production leads to a corresponding increase in income and a further increase in demand. II purchases slill exceed production firms lurlher increase production in the following quarter and so on.
• In short, in response to an increase in consumer spending, output doesn't jump to the new equilibrium, but rather increases over time from Y to Y'.
How long this adjustment takes depends on how and how often firms revise their production schedule. The more often firms adjust their production schedule, and the larger the response ol production to past increases in purchases, the faster the adjustment.
We will olten do in this book what we iust did here. Having looked at changes in equilibrium output, we will then describe informally how the economy moves from one equilibrium to the other. This will not only make the description ol what happens in the economy feel more realistic but will often reinforce your intuition about why the equilibrium changed.
We have focused in this section on increases in demand But the mechanism is symmetric— decreases in demand lead to decreases in output. In 1990-91 Australia had a serious recession output fell in three quarters), partly as the result ot a drop in government spending which led to a sharp decline in output. At the same time a recession happened in the United States, but tor diflerent reasons—a sudden drop in consumer confidence, leading to a large lall in output there. The US recession then adversely affected Australian exports to the United States, which deepened Australia's recession. The origins ol the 1990-91 recession in the United States arc examined in the focus box 'Consumer conlidence and the US recession ol 1990 91'.
CONSUMER CONFIDENCE AND THE US RECESSION OF 1990-91
In the third quarter of 1990. after the invasion of Kuwait by Iraq but before the beginning of the Persian Gulf War. US GDP growth turned negative, and remained negative for the following two quarters.This episode is known as the 1990-91 recession.
For a glimpse at the > longer list, go to the focus box 'Fiscal policy: What you have learned and where', in Chapter 27.
chapter 3
Column I ofTable I shows the size and the timing of the recession. It gives the change in GDP—in billions of 1992 dollars—for each quarter from the second quarter of 1990 to the second quarter of 1991. In 1990:3. 1990:4 and 1991:1, the change in GDP is negative.This is the 1990-91 recession.
Table 1 GDP. consumption, and forecast errors, 1990-91
(I) (2) (3) (4)
Change in Forecast error Forecast Index of consumer
Quarter real GDP for GDP error for CQ confidence
1990:2 1990:3 1990:4 19 -17 -23 105
-29 -57 -88 -1 90
-63 -37 61
1991:1 -31 -27 -30 65
1991.2 27 47 8 77
1 1
SOURCE Olivier Blanchard. 'Consumption and die recession of 1990-199 Г. American Economic Review, May 1993.
• Was the recession forecast by economists? The answer is 'no'. Column 2 gives the forecast error, the difference between the actual value of GDP and the value of GDP that had been forecast by economists one quarter earlier. A positive forecast error indicates that actual GDP turned out to be higher than was forecast: a negative forecast error indicates that actual GDP turned out to be lower than was forecast. As you can see, the forecast errors are negative during all three quarters of the recession.They are larger than the actual declines in GDP in each of the first two quarters of the recession.What this means is that, at the beginning of each of these two quarters, the forecasts were of positive GDP growth, while growth actually turned out to be negative.
• Where did these forecast errors come from! In terms of equation (3.8), which of the determinants of spending was the main culprit? Was it c0. or I. or G or T? Research looking at the evolution of each of the components of spending suggests that the main culprit, for the last two quarters of the recession, was an adverse shift in consumption, an unexpected decrease in c0. Forecast errors of c0 are given in column 3. There are two large negative errors for the last two quarters of the recession.
• A large decrease in CQ is a drop in consumption given disposable income.Why did consumption drop so much, given disposable income, in late 1990 and early 1991 ?The direct cause is shown in the last column of the table, which gives the value of the consumer confidence index.This index is calculated from a monthly survey of about 5,000 households: the survey asks consumers how confident they are about both current and future economic conditions, from job opportunities to their expected family income six months in the future. As you can see. there was a very large decrease in the index in the fourth quarter of 1990. Consumers lost confidence, leading them to cut consumption given disposable income, thus triggering the recession.
• This brings us to the last question: Why did consumers lose confidence in late 1990? Why did they become more pessimistic about the future? Even today, economists aren't sure. It is more than likely that this change in mood was related to the increasing probability of a war in the Middle East—a war that started in early 1991. after the beginning of the recession. People worried that the United States might become involved in a prolonged and costly war. They also worried that a war in the Middle East could lead to a large increase in oil prices and to a recession.The two previous large increases in oil prices in the 1970s had both been associated with recessions. Whatever the reason, the decrease in consumer confidence was a major factor behind the 1990-91 recession in the United States.
What about the 2001-02 slowdown in the United States where output growth fell to 0.3 per cent? First, things looked quite different from the 1990-91 recession. The slowdown came primarily from a decline in investment, rather than from a decline in consumption. Indeed, until September 11. consumer confidence and consumer spending had remained surprisingly high. One of the main worries, following September I I. was that consumer confidence would drop, leading to a decline in consumption and a further decline in output. Consumer confidence dropped, but by less than had been feared: the index, which was at I 10 in August, dropped to 85 in October—a much smaller decline than in 1990. By the end of 2002, US output growth had rebounded to 2.4 per cent. Second, was the US slowdown mirrored in Australia, as happened in the 1990-91
recession? Hardly. From 1992 to 1999. output in Australia grew at a consistent annual average of about 4 per cent, but dipped to just under 3 per cent for the next two years. Since that dip occurred in 2000 as well as 2001, it cannot be attributed to the feared fall in consumer confidence from the events of I I September 2001. In 2002. Australian output growth was again around 4 per cent, largely driven by real investment. Australia led the OECD in terms of output performance in those years.
3.4 INVESTMENT EQUALS SAVING: AN ALTERNATIVE WAY OF THINKING ABOUT GOODS—MARKET EQUILIBRIUM
Thus tar. we have been thinking ol equilibrium in the goods market in terms ol the equality ol the production and the demand for goods. An alternative—but equivalent—way ol thinking about equilibrium focuses instead on investment and saving. This is how John Maynard Keynes first articulated this model in 1936, in lhe General Theory of Employment. Interest ami Money.
• Let's start by looking at saving. By definition, private saving (S i, saving by consumers is equal to their disposable income minus their consumption:
5 s Yd - С
Llsing the definition of disposable income, we can rewrite private saving as income minus taxes minus consumption:
S = Y - T - С
• Now return to the equation lor equilibrium in the goods market. Production must be equal to demand, which, in turn is the sum ot consumption, investment and government spending:
Y - С 4 / + G
Subtract taxes (Tl from both sides and move consumption to the lelt side:
Y-T-C=I+G-T The left side ol this equation is simply private saving (S), so,
S / + С - T
Or equivalently.
I = S + (T - G) (3.10)
The term on the left is investment. The first term on the right is private saving. The second term is public saving—taxes minus government spending. If taxes exceed government spending, the govern¬ment is ainning a budget surplus—public saving is positive. II taxes are less than government spending, the government is running a budget deficit—public saving is negative.
Equation (3.10) gives us another way ol thinking about equilibrium in the goods market. It says that equilibrium in the goods market requires that investment equals saving—the sum ol private and public saving. This way ol looking at the equilibrium explains why the equilibrium condition for the goods market is called the IS relation, for 'Investment equals Saving'. What firms want to invest must be equal to what people and the government want to save.
©
To strengthen your intuition lor equation (3.10), think of an economy where there is only one person who has to decide how much to consume, invest and save—a Robinson Crusoe' economy. For Robinson Crusoe, the saving and the investment decisions are one and the same. What he invests (say, by keeping rabbits for reproduction, rather than having them for dinner he automatically saves. In a modern economy, however investment decisions are made by firms, whereas saving decisions are made by consumers and the government. In equilibrium, equation (3.10) tells us, all those decisions have lo be consistent—investment must be equal to saving.
To summarise: There are two equivalent ways ot stating the condition lor equilibrium in the goods market:
Production - Demand Investment Saving
Earlier, we characterised the equilibrium using the first condition, equation (3.6). We now do the same using the second condition, equation .3.10). The results will be the same, but the derivation will give you another way of thinking about the equilibrium.
• Note lirst that consumption unit saving decisions are one and the same. Given their disposable income, once consumers have chosen consumption, their saving is determined, and vice versa. The way we specified consumption behaviour implies that private saving is given by
5 = У - T - С
= Y - T - c0 - c,(Y - T)
Rearranging, we get
S= -c0 + (I —fj)(Y- T) (3.11)
• In the same way that we called с, the propensity to consume, we can call (I - r,) the propensity to save. The propensity to save tells us how much people save out ol an additional unit of income. The assumption we made earlier that the propensity lo consume (c,) is between zero and one implies that the propensity to save (I -ct) is also between zero and one. Private saving increases with disposable income, but by less than one dollar tor each additional dollar ot disposable income.
In equilibrium, investment must be equal to saving, the sum of private and public saving. Replacing private saving in equation (3.10) by its expression Irom above:
/ — -Co + (I -c,)(Y-T) + (T-G)
Solving tor output:
У = —'—LCo + Г + G - C,T1 (3.12)
I -C,
focus
/v Л
f box
Equation (3.12) is exactly the same as equation (3.8). This should come as no surprise. We are looking at the same equilibrium condition, just in a different way. This alternative way will prove uselul in various applications later in the book. I he focus box below looks at such an application, which was lirst emphasised by Keynes, and is often called the 'paradox of saving'
THE PARADOX OF SAVING
As we are growing up. we are told of the virtues of thrift.Those who spend all their income are condemned to end up poor.Those who save are promised a happy life. Similarly, governments tell us that an economy that saves is an economy that will grow strong and prosper.The model we have seen in this chapter, however, tells a different and surprising story.
Suppose that, at a given level of disposable income, consumers decide to save more. In other words, suppose consumers decrease c0. therefore decreasing consumption and increasing saving at a given level of disposable income. What happens to output and to saving?
Equation (3.12) makes clear that equilibrium output decreases. As people save more at their initial level of income, they decrease their consumption. But this decrease in consumption decreases demand, which decreases production.
IHhSIIOKl K'JN
chapter 3
Can we tell what happens to saving? Return to the equation for private saving, equation (3.1 I). (By assumption, there is no change in public saving, so saving and private saving move together.)
S = -co + (1 - с,)(У - T)
On the one hand. -Co is higher (less negative). Consumers are saving more at any level of income: this tends to increase saving. But, on the other hand, their income. У, is lower. This decreases saving. The net effect would seem to be ambiguous. In fact, we can tell which way it goes.
To see how, go back to equation (3.10).the equilibrium condition that investment and saving must be equal:
/ = S + (T - G)
By assumption, investment doesn't change: / = I. Nor does T or G. So, the equilibrium condition tells us that, in equilibrium, private saving. S, cannot change either. While people want to save more at a given level of income, income decreases by an amount such that saving is unchanged.
This means that attempts by people to save more lead both to a decline in output and to unchanged saving.This surprising pair of results is known as the paradox of saving (or the paradox of thrift).
So. should you forget the old wisdom? Should the government tell people to be less thrifty? No. The results of this simple model are of much relevance in the short run.The desire of consumers to save more led to the 1990-91 recession in the United States (as we saw in the focus box earlier in the chapter). But— as we will see later in the book when we look at the medium run and the long run—other mechanisms come into play over time, and an increase in the saving rate is likely to lead eventually to higher saving and higher income. A warning remains, however: policies that encourage saving may be good in the medium run and in the long run, but may lead to a recession in the short run.
3.5 IS THE GOVERNMENT OMNIPOTENT? A WARNING
the role of the government in general, and the succcsslul use ol liscal policy in particular, becomes increasingly difficult. Governments will never again have it so good as they had it in this chapter!
SUMMARY
1 hese are the facts you should remember about the components ol GDI':
• GDP is the sum ol consumption, plus investment, plus government spending, plus exports, minus imports plus inventory investment.
• Consumption (C) is the purchase of goods and services by consumers. Consumption is the largest component ot demand.
• Investment (/ is the sum of non-residential investment -the purchase of new factories and new machines by firms—and residential investment the purchase of new houses or apartments by people.
• Government spending (G) is the purchase ol goods and services by lederal. stale and local governments.
• F.xports (X) are purchases of Australian goods by foreigners. Imports (Ш arc purchases of foreign goods by Australian consumers, firms and the government.
• Inventory investment is the difference between production and purchases It can be positive or negative. This is what you should remember about our lirst model ol output determination:
• In the short run, demand determines production. Production is equal to income. And income determines demand.
• The consumption function shows how consumption depends on disposable income. The propensity to consume describes how much consumption increases lor a given increase in disposable income.
• Equilibrium output is the level of output at which production equals demand. In equilibrium, output equals autonomous spending times the multiplier. Autonomous spending is that part ol demand that doesn't depend on income. The multiplier is equal to I I — r, where r, is the propensity to consume.
• Increases in consumer confidence, investment demand or government spending, or decreases in taxes, all increase equilibrium output in the short run.
• An alternative way of stating the goods—market equilibrium condition is that investment must be equal to saving, the sum ol private and public saving, for this reason, the equilibrium condition is called the /5 relation (1 for investment, 5 lor saving I.
KEYTERMS
trade balance, 52 trade surplus, 52 trade deficit, 52 inventory investment, 52 identity, 52
disposable income (Yp), 53
consumption function, 53
behavioural equation, 53
linear relation, 53
parameter, 53
propensity to consume (cj), 53
• endogenous variables. 54
• exogenous variables, 54
• fiscal policy. 55
• equilibrium, 55
• equilibrium in the goods market, 55
• equilibrium condition. 55
• autonomous spending, 56
• balanced budget, 56
• multiplier, 56
• geometric series, 59
• econometrics. 59
dynamics, 60
forecast error, 6!
consumer confidence index, 61
• private saving (SI, 62
• public saving (7 -G), 62
• budget surplus, 62
• budget deficit, 62
• saving. 62
• IS relation, 62
• propensity to save, 63
• paradox of saving, 64
QUESTIONS AND PROBLEMS
Quick check
1. Using the information in this chapter, label each of the following statements 'true', 'false' or 'uncertain'. Explain briefly.
a. The largest component of ( .1)1 is consumption.
b. Government spending, including transfers, was equal to 22 per cent ot GDP in 2008 in Australia.
c. The propensity to consume has to be positive, but otherwise it can take on any positive value.
d. fiscal policy describes the choice ol government spending and taxes, and is treated as exogenous in our goods market model.
e. The equilibrium condition lor the goods market states that consumption equals output.
f. An increase ot one unit in government spending leads to an increase of one unit in equilibrium output.
g. A decrease in the propensity to consume leads to an increase in output.
2. Suppose that the economy is characterised by the following behavioural equations:
С = 160
I 150
G - 150
T = 100
Solve for
a. Equilibrium GDP (У)
b. Disposable income (Yp)
c. Consumption spending (C)
3. For the economy in problem 2,
a. Solve lor equilibrium output. Calculate total demand. Is it equal lo production? Explain.
b. Assume that G is now equal to 1 10. Solve for equilibrium output. Calculate total demand. Is il equal to production? Explain.
c. Assume that CJ is equal to I 10, so output is given by your answer to (bi. Calculate private plus public saving. Is ii equal to investment? Explain.
Dig deeper
4. The balanced budget multiplier
For both political and macroeconomic reasons, governments are often reluctant to run budget deficits. Here, we examine whether policy changes in G and Г that maintain a balanced budget are macroeconomically neutral. Put another way. we examine whether it is possible to affect output through changes in G and T so that the government budget remains balanced. Start from equation 13.7).
a. By how much docs Y increase when G increases by one unit?
b. By how much does Y decrease when Г increases by one unit?
c. Why arc your answers to (a) and (b) different?
Suppose that lhe economy starts with a balanced budget: Г - С!, if the increase in G is equal to the increase in T, the budget remains in balance. Let us now calculate the balanced budget multiplier.
d. Suppose that both G and 7 increase by one unit. Using your answers to (ai and (b , what is the change in equilibrium GDI'? Are balanced budget changes in G and T macroeconomically neutral?
e. How does the specific value of the propensity to consume affect your answer to (dl? Why?
5. Automatic stabilisers
St) far in this chapter we have been assuming that the fiscal policy variables G and T 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. In this problem we examine how this automatic response of taxes can help reduce the impact of changes in autonomous spending on output.
Consider the following behavioural equations: С = C„ + C| Y;> T= t0 + f,Y Yn = Y-T
C", and I are both constant.
Assume that t. is between zero and one.
a. Solve for equilibrium output.
b. What is the multiplier? Does the economy respond more to changes in autonomous spending when fj is zero or when I, is positive? Explain.
c. Why is liscal policy in this case called an automatic stabiliser'?
6. Balanced budget versus automatic stabilisers
It is often argued that a balanced budget amendment would actually be destabilising. To understand this argument, consider the economy o f problem 5.
a. Solve lor equilibrium output.
b. Solve for taxes in equilibrium.
Suppose that the government starts with a balanced budget and that there is a drop in r„.
c. What happens to Y? What happens to taxes?
d. Suppose that the government cuts spending in order to keep the budget balanced. What will be the effect on Y? Does the cut in spending required to balance the budget counteract or reinforce the effect of the drop in c() on output? (Don't do the algebra. Use your intuition and give the answer in words.)
7. Taxes and transfers
Recall that we define taxes, T, as net of transfers. In other words, T = Taxes — transfer payments
a. Suppose that the government increases transfer payments to private households, hut these transfer payments are not financed by tax increases. Instead, the government borrows to pay lor the transfer payments. Show in a diagram (similar to Figure 3.2) how this policy affects equilibrium output. Explain.
b. Suppose instead that the government pays for the increase in transfer payments with an equivalent increase in taxes. How does the increase in transler payments allect equilibrium output in this case?
c. Now suppose that the population includes two kinds of people: those with high propensity to consume and those with low propensity to consume. Suppose the transler policy increases taxes on those with low propensity to consume to pay for transfers to people with high propensity to consume. How does this policy affect equilibrium output?
d. How do you think the propensity to consume might van,' across individuals according to incomer In other words, how do you think the propensity to consume compares lor people with high income and people with low income? Explain. Given your answer, do you think tax cuts will be more effective at stimulating output when thev arc directed toward high-income or towards low-income taxpayers?
8. Investment and income
This problem examines the implications of allowing investment to depend on output. Chapter 5 carries this analysis much further and introduces an essential relation—the effect of the interest rate on investment—not examined in this problem.
a. Suppose the economy is characterised by the following behavioural equations:
С = c0 + с Yn YD = Y-T I - К + /'.V
Government spending and taxes arc constant. Note that investment now increases with output. (Chapter 5 discusses the reasons lor this relation.) Solve for equilibrium output.
b. What is the value ol the multiplier? How does the relation between investment and output alfect the value of the multiplier? for the multiplier to be positive what condition must satislv? Explain your answers.
c. Suppose that the parameter b„, sometimes called business confidence. increases. How will equilibrium output he affected? Will investment change by more or less than the change in /',,? Why? What will happen to national saving?
Explore further
9. The paradox of saving revisited
Voli should be able to complete this question without doing any algebra, although you may find making a diagram helpful for part la I. For this problem, you do not need to calculate the magnitudes of changes in economic variables—only the direction of change.
a. Consider the economy described in problem 8. Suppose that consumers decide to consume less land therefore to save more for any given amount of disposable income. Specifically, assume that consumer confidence if,,' talis. What will happen to output?
b. As a result ol the effcct on output you determined in part a1, what will happen to investment? What will happen to public saving? What will happen to private saving? Explain, ii/i'iil: Consider the saving equals investment characterisation of equilibrium, i What is the effect on consumption?
c. Suppose that consumers had decided to increase consumption expenditure, so that ty. bad increased. What would have been the effcct on output, investment and private saving in this case? Explain. What would have been the ellect on consumption?
d. Comment on the following logic: 'When output is too low, what is needed is an increase in demand for goods and services. Investment is one component ol demand, and saving equals investment. Therefore, il the government could just convince households to attempt to save- more, then investment, the demand lor goods and services, and output, would increase.
Output is not the only variable that a ffects investment. As we develop our model of the economy, we will revisit the paradox of saving in future chapter problems.
We invite you to visit the Blanchard-Sheen page on the Pearson Australia website at
www.pearson.com.au/highered/blanchardsheen3e
for many World Wide Web exercises relating to issues similar to those in this chapter.
B
arely a day goes by without the media speculating whether the Reserve (short for the Reserve Bank of Australia, or RBA) or the Fed (short for the Federal Reserve Bank, the central bank of the United States) is going to change the interest rate, and what the change is likely to do to the economy. Our central bankers have become the most powerful of policy-makers, and an analysis of the performance of the economy cannot ignore them. The model of economic activity developed in Chapter 3 didn't include the interest rate, so there was no role for our central bankers.This was a strong simplification: it is time to relax it.This requires that we take two steps. First, we must look at what determines the interest rate, and at the role the central bank plays in this determination—the topic of this chapter. Second, we must look at how the interest rate affects demand and output—the topic of the next chapter. The chapter has four sections:
• Section 4.1 looks at the demand for money.
• Section 4.2 assumes that the central bank directly controls the supply of money, and shows how the interest rate is determined by the market equilibrium condition that the demand for money be equal to the supply of money.
• Section 4.3 (an optional section) introduces banks as suppliers of money, and revisits the market determination of the interest rate and the role of the central bank.
Make sure you see the difference between the decision about how much to save (a decision that determines how wealth changes over time) and the decision about how to allocate a given stock of wealth between money and 4 bonds.
• Section 4.4 (also an optional section) presents two alternative ways to think about the equilibrium. One focuses on the interbank overnight cash market in Australia, which is equivalent to the federal funds market in the United States.The other focuses on the money multiplier.
4.1 THE DEMAND FOR MONEY
This section looks at the determinants of the demand for money. (A warning before we start: Words such as money or wealth have very specilic meanings in economics, often not the same meanings as in everyday conversations. The purpose ol the focus box Semantic traps: Money, income and wealth' is to help you avoid some of these traps. Read it carefully, and come back to it once in a while.)
Suppose, as a result of having steadily saved part ol your income in the past, your linanciai wealth today is $50,000. You may intend to keep saving in the future and lo increase your wealth lurther, but its value today is given. The only choice you can make today is bow lo allocaic this $50,000 between money and bonds:
CHAPTER О
Financial Markets
• Money, which you can use for transactions, pays no interest.
In reality, there are two types of money: currency—coins and notes—and current accounts—those hank deposits that you can use for direct payment (via cheques or EFTPOS). The distinction between the two will he important when we look at the supply ol money. For the moment, it doesn't matter. • Bonds pay a positive interest rale, /', but they cannot he used lor transactions.
In reality, there are many types ol bonds, each associated with a spccilic interest rate for the time being, we will ignore this aspect of reality and assume there is just one type of bond, so i is the interest rate.
Think ol buying or selling bonds as implying some cost—for example, a phone call to a broker and the payment of a transaction fee. How much of your $50,00(1 should you hold in money, and how much in bonds?
Holding all your wealth in the form of money is clearly very convenient. It avoids the need to call a broker or pay transaction fees. But it also means receiving no interest income.
Holding all your wealth in the form ol bonds implies receiving interest on all your wealth, but having to call your broker whenever you need money to take the bus or pay for a cup of coffee is an inconvenient way of going through life!
Every day we use money to denote many things. We use it as a synonym for income:'He's into making money'. We use it as a synonym for wealth:'She has a lot of money'. In economics, you must be more careful. Here is a basic guide to some terms and their precise meanings in economics.
Income is what you earn from working, plus what you receive in interest and dividends. It is a flow: that is, it is expressed per unit of time—weekly income, monthly income or yearly income. The fabulously rich J. Paul Getty was once asked what his income was. Getty answered. '$ 1.000.' What he meant, but didn't say, was,'$1,000 per minute.'
Saving is that part of after-tax income that isn't spent. It is also a flow. If you save 10 per cent of your income and your income is $3,000 per month, then you save $300 per month. Savings (plural) is sometimes used as a synonym for wealth—the value of what you have accumulated over time. To avoid confusion, we won't use 'savings' (plural) in this book.
Your financial wealth, or simply wealth, is the value of all your financial assets minus all your financial liabilities. In contrast to income and saving, which are flow variables, financial wealth is a stock variable. It is the value of wealth at a given moment in time.
At a given moment in time, you cannot change the total amount of your financial wealth.You can change it only over time, as you save or dissave, or as the value of your assets change. But you can change the composition of your wealth.You can, for example, decide to pay back part of your mortgage by reducing your bank account.This leads to a decrease in your liabilities (a smaller mortgage) and a corresponding decrease in your assets (a smaller bank balance); but it doesn't change your wealth.
Financial assets that can be used directly to buy goods are called money. Money includes currency and all bank accounts that you can use to make direct payments, either with cheques or by EFTPOS (electronic funds transfer at the point of sale). Money is also a stock. Somebody can have a large wealth but small money holdings—for example. $ I million worth of stocks, but only $500 in her bank account. Or somebody can have a large income but small money holdings—for example, be paid $10,000 a month, but have a very small positive balance at the bank.
Investment is a term economists reserve for the purchase of new capital goods, from machines to plants to office buildings. When you want to talk about the purchase of shares or other financial assets, you should call it financial investment.
Learn how to be economically correct:
Don't say 'Kylie is making a lot of money': say 'Kylie receives a high income'.
We will abandon this ► assumption and look at a large menu of interest rates when we focus on the role of expectations, starting in Chapter 15. We will also introduce stocks and housing.
Don't say 'Kerry has a lot of money': say 'Kerry is very wealthy'.
It is clear that you should hold both money and bonds. In what proportions should you do so? This will depend mainly on two variables:
• Your level of transactions. You want to have enough money, on average, to avoid having to sell bonds to get money too often. Say that you typically spend $3,000 a month. You may want to have on average, say, two months worth ol spending on hand, or $6,000 in money, and the rest, $50,000 - $6,000 - $44,000, in bonds. Il, instead, you typically spend $4,000 a month, you may want to have $8,000 in money and only $42,000 in bonds.
• The interest rate on bonds. The only reason to hold any of your wealth in bonds is that they pay interest. Il bonds paid no interest, you would hold all ol your wealth in money. Bonds and money would pay the same interest rate namely, zero), and money, which can be used lor transactions, would therefore be more convenient.
The higher the interest rate, the more you will be willing to incur the hassle and the costs associated with buying and selling bonds. Il the interest rale is very high, you may decide to squeeze your money holdings to an average ot only two weeks' worth of spending, or $1,500 (assuming your monthly spending is $3,000). This way. you will be able to keep, on average, $48,500 in bonds, getting more interest as a result.
I.et's make this last point more concrete. Most ot you probably don't hold bonds, lew of you have a broker. But many ol you hold bonds indirectly, through perhaps a term deposit account. Il you try and cash this in early, you lose the interest. The bank pays an interest rale on term deposits close to the interest rate on government bonds—the dillerence coming Irom the administrative costs of running lhe bank and from its profit margin.
In the late 1980s, with the interest rate reaching 17 per cent per year, many people who had previously kept all their financial wealth in their bank deposit account which paid little or no interest i realised how much interest they could earn by holding part ol their financial wealth in a term deposit. Term deposits became very popular. Since then, however the interest rate has decreased. In 2006. the interest rate paid on term deposits was typically about 3 per cent. This is better than zero—lhe rate paid on current account deposits—but is much less attractive than the rate in the 1980s. As a result people arc now less careful about putting as much as they can in their term deposit. Put another way, lor a given level of transactions, people now keep more in their current account than they did in the 1980s.
Deriving the demand for money
Let's move from this discussion to an equation describing the demand for money.
Denote ihe amount or money people want to hold—their demand lor money—by Л1'1. (The superscript d stands for demand. The demand for money lor the economy as a whole is just the sum of all the individual demands lor money. Thus, money demand lor the economy as a whole depends on the overall level ol transactions in the economy and on the interest rate. The overall level ol transactions in lhe economy is hard to measure, but it is likely to be roughly proportional to nominal income. If nominal income increases by 10 per cent, it is reasonable lo think that the amount ol transactions in the economy also increases by roughly 10 per cent. So we can write lhe relation between the demand for money, nominal income and the interest rale as
MJ = $YL(f) (4.1)
(-)
where $Y denotes nominal income. Read this equation in the following way: The demand for money, M1', is equal to nominal income, SY, times
What matters here is nominal income— income in dollars—not real income. If real income doesn't change but prices double, leading to a doubling of nominal income, people will need to hold twice as much money to buy the same consumption 4 basket.
Equation (4.1) summarises what we have discussed so far: • First, the demand for money increases in proportion lo nominal income. If nominal income doubles, increasing from $Y to $2Y the demand lor money also doubles, increasing Irom $YLHi to $2YL(t).
• Second, the demand for money depends negatively on the interest rate. This is captured by the function Ltit and the negative sign underneath: an increase in the interest rate decreases the demand (or money.
The relation between the demand for money, nominal income and the interest rate implied by equation (4.1) is represented in Figure 4.1. I he interest rate, i. is measured on the vertical axis. Money, M, is measured on the horizontal axis.
The relation between the demand lor money and the interest rate for a given level of nominal income is represented by the AI" curve. The curve is downward sloping: the lower the interest rate (the lower is /), the higher the amount ot money people want to hold ithe higher is M).
For a given interest rate, an increase in nominal income increases the demand lor money. In other words, an increase in nominal income shilts the demand tor money to the right, from XI'1 to MJ'. For example, at interest rate i, an increase in nominal income tor $Y to $Y' increases the demand for money from M to M'.
Figure 4.1 The demand for money
M"' (for
$y > $Y)
Md (for nominal income $У)
M M
Money, /И
box
For a given level of nominal income, a lower interest rote increases the demand for money. At a given interest rate, an increase in nomina1 income shifts the demand for money to the right.
THE DEMAND FOR MONEY AND THE INTEREST RATE: THE EVIDENCE
How well does equation (4.1) fit the facts? In particular, how much does the demand for money respond to changes in the interest rate? To get at the answer, first divide both sides of equation (4.1) by $Y:
g - m
The term on the left side of the equation is the ratio of money demand to nominal income—in other words, how much money people want to hold in relation to their income. Because L(i) is a decreasing function of the interest rate, i, this equation says:
• When the interest rate is high, then L{i) is low and the ratio of money to nominal income should be low.
• When the interest rate is low. then L(i) is high and the ratio of money to nominal income should be high.
So, if equation (4.1)—and. by implication, this equation—is a good description of reality, we should observe an inverse relation between the ratio of money to nominal income and the interest rate. This provides the motivation for Figure I. which plots both the ratio of money to nominal income and the interest rate against time, for the period 1960-2008 in Australia.
The ratio of money to nominal income is constructed as follows. Money, M, is the sum of currency and bank current account deposits that can be used directly for payments.This measure of money is called Ml. Nominal income is measured by nominal GDP. $Y.The interest rate, /', is the average interest rate paid by ten-year government bonds during each year.
Figure I suggests two main issues:
• The first is that there was a large decline in the ratio of money to nominal income from I960 to 1985, and a return to its value in the 1960s by 2002.The interest rate was roughly the same from 2002 to 2008 as it was in the 1960s.
Economists sometimes refer to the inverse of the ratio of money to nominal income—that is. to the ratio of nominal income to money—as the velocity of money. The use of the word velocity comes from the intuitive idea that, when the ratio of nominal income to money is higher, the number of transactions
Figure I
r 0.25
16 -,
Interest rate (i)
0.23
- 0.21
-0.19 g
I II I I
1980
0.17
0.15
0.13
50
p
i2 o' о
3 о
0.11
0.09
TTTTT
1970
TTTTT
1975
I I I I I I l l l l l l l l l l l l l II l l l
I98S 1990 1995 2000 2005
i i i i
1965
I960
The ratio of money to nominal income and the interest rate in Australia. 1960-2008
©
The ratio of money (Ml) to nominal income decreased from 1960 to 198S. and increased from 1985 to 2002. (The spike in 2002 was a statistical artefact due to a change in the way banks measured their deposits.I The interest rate (ten-year Treasury bonds) moved in the opposite direction. SOURCE: RBA G12. D03Hist. ABS 1364 Table 29
for a given quantity of money is higher, and it must be the case that money is changing hands faster; in other words, the velocity of money is higher. Therefore, another, equivalent, way of stating the first characteristic of Figure I is that the velocity of money has increased from about 4.6 (1/0.217) in I960 to about 9.8 (1/0.102) in 1985 and back down to 4.7 (1/0.213) in 2007.
Why did velocity more than double from 1960 to 1985 and then more than halve? The reasons aren't hard to find. First, the interest rate can explain much of what has happened. Second, dramatic financial innovations have occurred in financial markets in the last forty years.
The interest rate went up from just over 6 per cent in I960 to a peak of 14.8 per cent in 1985.Then it came down to 6.2 per cent in 2007.This means that our liquidity function, L(i). would have fallen and risen in much the same way as MI$Y in Figure I. Since velocity is the inverse of Ц1) in money market equilibrium, it would have done the reverse. But that isn't all the story. Interest rates didn't really take off until 1973, and yet velocity was rapidly increasing from I960 to 1973. So. there must be a second explanation that isn't obvious from the figure.
Many innovations in financial markets have made it possible to hold lower money balances for a given amount of transactions. Perhaps the most important development here is the increased use of credit cards. At first glance, credit cards would appear to be money. When we go to a department store, aren't we asked whether we want to pay with cash, debit card or credit card? But. despite what they may seem, credit cards aren't money.You actually don't pay when you use your credit card at the store; you pay when you receive your bill and send your monthly payment. A credit card offers you a type of loan, and we wouldn't want to include loans in our definition of money. What credit cards allow you to do is to concentrate many of your payments in one day. and thus to decrease the average amount of money you need to have during the rest of the month. (Some credit cards also allow you to defer payment beyond a minimum amount, and thus to borrow, often at a high interest rate.This is a separate service, and not what is relevant here. We are interested in how a credit card helps us to economise on needing money.) You would expect the introduction of credit cards to reduce money demand in relation to nominal income over time. Figure I suggests that this has been the case, from 1960 to 1973.
Why don't we see the same downward trend in M!$Y in the 1980s and 1990s? As we have already said, the main reason is that interest rates fell, stimulating money demand. Another reason is probably because of EFTPOS facilities, which emerged in the mid-1980s and are ubiquitous now in Australia. EFTPOS has made it easier for us to use money (that is, our bank accounts) with a debit card, and thus for many transactions we don't need to resort to the credit card. Therefore. EFTPOS raises the demand for rroney, cancelling out the effect of credit cards on velocity (and its inverse M/$Y). Or, to put it another way, EFTPOS lowers transaction costs and increases the efficiency of the payments system, so that the growth in nominal income (and hence transactions) can be supported by a more than proportional growth in money.
• The second issue follows from the first—we can conclude that there is a negative relation between year- to-year movements in the ratio of money to nominal income and year-to-year movements in the interest rate. A good way to look at year-to-year movements is with a scatter diagram.
Figure 2 plots the change in the ratio of money to nominal income versus the change in the interest rate from year to year. Changes in the interest rate are measured on the vertical axis. Changes in the ratio of money to nominal income are measured on the horizontal axis. Each point (shown as a square) in the figure corresponds to a given year. The vertical and horizontal lines give the mean values of the change in the ratio and in the interest rate for the period l960-2008.The figure shows a negative relation between year-to-year changes in the interest rate and changes in the ratio. Note that most of the points lie either in the northwest quadrant (increases in the interest rate, decreases in the ratio) or the southeast quadrant (decreases in the interest rate, increases in the ratio). But notice that there are quite a few points in the northeast quadrant, three of which are from 2006 to 2008—these probably reflect the developing 'sub-prime' crisis which led to higher interest rates as a response to increased global financial risk (this issue will be discussed more in Chapter 22). So the relation isn't tight, but when we draw a line that best fits the cloud of points, it is clearly downward sloping, as predicted by our money demand equation.
Figure 2
Changes in the interest rate versus changes in the ratio of money to nominal income in Australia, 1960-2008
4.0-,
3.0 -
2.0-
1.0 -
"rxHT"
Па 3d-.D
1 an
0.0-
d3 —
ED
□ □ □
с re x U
-2.0-
-r
-0.03 -0.02 -0.01 0 0.01
-3.0- -0.03
т
—I—
0.02
—I
0.03
Changes in the ratio of money to annual income (%)
Increases in (he interest rate have typically been associated with a decrease in the ratio of money to nominal income, and decreases in the interest rate with on increase in that ratio.
4.2 MONEY MARKET EQUILIBRIUM AND THE INTEREST RATE: I
Having looked at the demand for money, we now look at the supply ol money, and then at the equilibrium in the money market.
In reality, there are two suppliers ol money: current account deposits useful tor payments arc- supplied by banks,- and currency is supplied by the central bank. In this section, we will assume that people hold only currency as money, so all money is currency supplied by the central bank. In the next section, we will reintroduce deposits, and look at the role of banks. Introducing banks makes the discussion more realistic. But it also makes the mechanics ol money supply more complicated, and it is Ix-tter to build the intuition in two steps.
Money demand, money supply and the equilibrium interest rate
Suppose the central bank decides to supply an amount ol money equal to ,Vt, so M' = Л1. The superscript s stands for supply. I Let's leave aside for the moment the issue ot how the central bank determines the amount of money in the economy. We will return to it shortly.
Equilibrium in financial markets requires that money supply be equal to money demand, that Ms - Md. Then. using Ms - M, and equation (4.11 for money demand, the equilibrium condition is
14.2)
Money supply = Money demand
M - $YL(i)
Equation (4.2) tells us that the interest rate / must be such that, given their income 5Y, people arc willing to hold an amount ot money equal to the existing money supply M. This equilibrium relation is called the LM relation.
Throughout this section, mcney stands for central 4 bank money, or currency.
As for the IS relation, the name of the LM relation is more than fifty years old.The letter L stands for 'liquidity'. Economists use liquidity as a measure of how easily an asset can he exchanged for money. Money is fully liquid. other assets less so. we can think of the demand for money as a demand ^ for liquidity. The letter M stands for money.The demand for liquidity must equal the supply of money.
This equilibrium condition is represented graphically in figure -1.2. As in Figure 4.1, money is measured on the horizontal axis, and the interest rate is measured on the vertical axis. The demand for money, Mdrawn for a given level ol nominal income $Y, is downward sloping—a higher interest rate
Figure 4.2
Money supply M'
Money market equilibrium and the interest rate
\
01
я
\
\
с
©
\
\
V
S
M
Money, /И
Figure 4.3 The effects of an increase in nominal income on the interest rate
M
Money, /И
($v >$Y)
An increase in nominal income leads to an increase in the interest rate.
Figure 4.4 The effects of an increase in the money supply on the interest rate
An increase in the money supply lecds to a decrease in the interest rate.
Assume that the central hank changes the supply of money hy buying or selling bonds in the bonds market. To increase the amount of money in the economy, it buys bonds and pays for them by creating money. Го decrease the amount ol money in the economy, it sells bonds and removes from circulation the money it receives in exchange for the bonds. Such operations are called open-market operations, so-called because they take place in the open market' lor bonds. They are the standard method central banks use to change the money stock in modern economies.
The balance sheet, or statement of financial position, of the central bank is given in Figure 4.5. The assets ol the central bank arc the bonds it holds in its portfolio. Its liabilities arc the stock ot money in the economy. Open-market operations lead to equal changes in assets and liabilities.
If the central bank buys, say, $1 million worth of bonds, the amount ol bonds it holds is higher by $1 million, and so is the amount of money in the economy. Such an operation is called an expansionary open-market operation, because the central bank increases (expands) the supply of money (which, we will see. reduces the interest rate I.
If the central bank sells $1 million worth of bonds, both the amount of bonds held by the central bank and the amount of money in the economy are lower by $1 million. Such an operation is called a contractionary open-market operation because the central bank decreases (contracts: the supply ot money (which, we will see, raises the interest rate).
The balance sheet, or ► statement of financial position, of a bank (or firm, or individual) is a list of its assets and I abilities at a point in time. The assets are the sum of what the bank owns and what is owed to the bank at that time. The liabilities are what the bank owes to others, also at that time.
chapter 4
One more step is needed before we can describe the ellects ol open-market operations. We have focused so lar on the interest rale on bonds. In fact, what is determined in bond markets isn't interest rales but bond prices,- ihe interest rate on a bond can then be interred Irom the price ol lhe bond. Understanding the relation between the interest rate and the price ol a bond will prove useful both here and later in the book.
The interest rate is what ► you get for the bond a year from now ($100) minus what you pay for the bond today (SP6). divided by the price of the bond today ($Pt).
I =
(a) Balance sheet
Assets Liabilities
Bonds Money (currency)
(b) The effects of an expansionary
open-market operation
Assets Liabilities
Change in bond Change in money
holdings: stock:
+$l million +$l million
Figure 4.5 The balance sheet of the central bank, and the effects of an expansionary open-market operation
The assets of the central bank are the bonds it holds. The liabilities are the stock of money in the economy. An open-market operation in which the central bank buys bonds and issues money increases both assets and liabilities by the same amount
• Suppose the bonds in our economy are one-year bonds bonds that promise a payment ol a given number of dollars, say, $100, a year hence. In Australia, such bonds, when issued by the government and promising payment in a year or less, arc called Treasury bills or T-bills. Let the price ol a bond loday be $Рц, where the subscript 11 stands for 'bond'. It you buy the bond today and hold il lor a year, lhe rate of return on holding the bond lor a year is $100 - $PHi/$PB. Therelore. the interest rate on the bond is given by
$100 - $Pti $PB
If $Р,( is $95, the interest rate equals $5/$95 = 0.053, or 5.3 per cent per year. If $Pn is $90, the interest rate is III per cent per year. The higher the price of lhe hand, the lower the interest rale.
F.quivalently, il we are given the interest rale, we can inler ihe price of the bond. Reorganising the formula above, the price today of a one-vcar bond paying $100 a year from today is given by
$100 I + I
The price of the bond today is equal to the final payment divided by I plus the interest rate. II the interest rate is positive the price of the bond is less than the final payment. The higher the interest rate, the lower the price today. When newspapers write that bond markets went up today', they mean that the prici'> of 'ootids went up, and therefore that interest rates went down.
We are now ready to return to the effects ol an open-market operation. Consider lirst an expans¬ionary open-market operation. in which the central bank buys bonds in the bond market and pays lor them by creating money. As the central bank buys bonds, the demand lor bonds goes up, increasing the price ol bonds. F.quivalently, the interest rate on bonds goes down. Thus, if the central bank wants interest rates to go down, it can conduct expansionary open-market operations. If, instead, the central bank decreases the supply ol money— a contractionary open market operation—it sells bonds in the bond market. This leads to a decrease in their price and an increase in the interest rate. Thus, if the central bank wants interest rates to go up, it can conduct contractionary open-market operations. To summarise:
• The interest rate is determined by the equality ot the supply ol money and the demand for money.
• By changing the supply ot money, the central bank can allect the interest rale.
• The central bank changes the supply ol money through open-market operations, which arc- purchases or sales of bonds tor money.
• Open-market operations in which the central bank increases the money supply by buying bonds lead to an increase in the price of bonds—equivalentlv. a decrease in the interest rate.
• Open-market operations in which the central bank decreases the money supply by selling bonds lead to a decrease in the price ot bonds—equivalentlv, an increase in the interest rale.
4 The complication The short-term interest rate—the rate direcdy affected by monetary policy—isn't the only interest rate in the economy, and isn't the only interest rate that affects spending. The determination of other interest rates and asset prices (such as stock prices) is the topic of Chapter 15.
Let us take up two more issues before moving on.
Choosing money or choosing the interest rate?
We have described the central bank as choosing the money supply and letting the interest rate be determined at the point where money supply equals money demand. Instead, we could have described the central bank as choosing the interest rate and then adjusting the money supply so as to achieve this interest rate.
To see this, return to Figure 4.4. Figure 4.4 showed the effect ol a decision by the central bank to increase the money supply Irom ЛГ to AI', causing the interest rate to lall Irom i to i". However, we could have described the figure in terms of the central bank's decision to lower the interest rate from i to i' by increasing the money supply from ЛГ to At1'.
Why is it useful to think about choosing the interest rale? Because ibis is what modern central banks, including the RBA and the Fed, typically do. They typically think about the interest rate they want to achieve and then move the money supply so as to achieve it. This is why, when you listen to the news, you don't hear, The RBA decided to increase the money supply today. Instead, you hear, The RBA decided to decrease the interest rate today. The way the RBA did it was by increasing the money supply appropriately.
We will take this issue up again in the next chapter when we discuss the policy mix.
Money, bonds and other assets
SPn =
4 In Japan today, the one-year interest rate is (nearly) equal to zero. If a one-year Japanese government bond promises 100 yen in one year, for what price will it sell today!
We have been looking at an economy with only two assets, money and bonds. This is obviously a much simplified version of actual economies with their many financial assets and many financial markets. But, as you will see in later chapters, the basic lessons we have just seen apply very generally. The only
change we will have to make is to replace 'interest rate' in our conclusions hy 'short-term interest rate. You will see that the short-term interest rate is determined hy the condition that money supply equals money demand, the central hank can, through open-market operations, change the short-term interest rate,- and open-market operations are indeed the basic tool used by most modern central banks, including the RBA, to determine the desired short-term interest rate.
You can skip the next ► two sections and still go through most of the arguments in the rest of the book. In case you do so. let us give you the bottom line—even in this more complicated case, the central bank can, by changing the amount of central bank money, affect the interest rate.
As always, this ► description is a simplification. Banks have liabilities other than current account deposits, and are engaged in more activities than holding bonds or making loans. But these complications aren't relevant here.
2.
There is one dimension, however, in which our model must be extended. We have assumed that all money was currency, supplied by the central bank. In the real world, money includes not only currency but also current account deposits useful for payments. These deposits are supplied not by the central bank but by (private) banks. How the presence of banks changes our conclusions is the topic of the next two sections.
Комментариев нет:
Отправить комментарий