Tucking in! Consumption

Consumption C is the largest component of aggregate demand. In the U.S. it accounts for more than two-thirds of output. Essentially, around 68 percent of everything produced in the United States ends up being consumed by households in one way or another. So whether you’re slurping a soft drink or sipping a hard one, chomping on a chocolate bar or gobbling some grapes, you’re contributing to the country’s consumption.

Consumption doesn’t just include things you literally consume such as food and drink (and, if you’re Homer Simpson, flowers). All goods and services that households purchase are included in consumption. So the flat-screen TV you got for Christmas is consumption. That Uber ride you bought last week — that’s consumption. Even the rent you pay for your tiny overpriced New York apartment counts as the consumption of “housing services.”

Basically, if you buy something — and you aren’t an entrepreneur buying something for your business — it’s probably consumption.

Consumption function

The amount that households choose to consume depends on a number of different variables.

The most important is their disposable income, which is household income after tax. Total disposable income in an economy equals Y – T, where Y is total income (also equal to production or GDP) and T is total taxes households pay to the government.

You can express the relationship between consumption and disposable income mathematically with a consumption function:

In words, this expression says that consumption C is a function of disposable income Y – T. Specifying the consumption function explicitly as a linear relationship shows how consumption varies with disposable income:

c₀ is called the level of autonomous consumption, that is, consumption independent of the level of disposable income. c₁ is called the marginal propensity to consume and tells you how much consumption increases when disposable income increases by $1. It’s a number between 0 and 1.

Figure 10-1 shows a graph of the consumption function. The line intersects the y-axis at c₀, and the slope of the line is equal to c₁. If autonomous consumption increases, this shifts the entire consumption function upwards, whereas an increase in the marginal propensity to consume makes the consumption function steeper.

You may be wondering, how can c₀ be positive? How can households consume even if they have no disposable income, that is, if Y – T = 0? Good question. This is one of those places where the short-run model doesn’t quite hook up with our long-run analysis. A more complete treatment would recognize that when people decide how much they want to consume today, they’re simultaneously deciding how much they want to save because saving is what’s left of disposable income after consumption. And people save today to add to their consumption tomorrow. That’s what pension plans are all about.

So, a complete story would see the consumption choice this period as part of a plan for consumption now and in the future. This approach would allow us to recognize that current income is not the fundamental constraint on current consumption. People can consume even when they don’t have any current income by: a) borrowing against future income, or b) using their savings from earlier periods, what we call household wealth. Young people do a lot of the former (paying for college and so on) and older folk do a lot of the latter (living off their pensions, 401Ks, IRAs, savings, and so on in retirement). Combining current disposable income Y – T with the autonomous consumption term c₀, though is a mathematically simple way to capture these effects. It’s good for describing short-run behavior even if it’s not a complete representation of the underlying economics.

The consumption function makes clear that any increase in income Y raises consumption C, but by less than the increase in income. Any decrease in taxes T also increases consumption by less than the amount by which taxes fall. The reason is the same in both cases. The marginal propensity to consume is less than 1. At the margin, consumption changes less than disposable income.

Tooling up for business: Investment

Domestic private investment I is the third-largest component of aggregate demand, accounting for roughly 15 percent of GDP in the U.S. It is also the most volatile component of GDP, so this percentage can vary substantially. Typically, during a boom (a period of above-average economic growth) investment can rise sharply, whereas during a bust (a period of below-average or negative economic growth) investment can fall sharply.

Investment is one of those words that economists use substantially differently from the way laypeople use it. To the latter, investment often refers to the purchase of shares in a company or buying a house in the hope that it will appreciate in value. To an economist, investment is the purchase of new capital goods, which are goods used to produce other goods and services in the future.

Examples of capital goods include machines, computers, buildings, and so on. Firms carry out most capital goods investment — after all, they’re trying to make stuff to sell to people, and in order to make stuff, they need capital goods. For this same reason, adding to inventory is another form of investment. Firms can meet future demands by selling this inventory. (Turn to Chapter 2 for further discussion on what economists mean by the terms investment and capital.)

There is one case where economists count purchasing a house as investment: buying a newly built house (more precisely they call it residential investment). The idea is that purchasing a newly built house reflects a demand for new (housing) capital goods. If you think about it, a house provides you with a stream of “housing services” in the future, a lot like a new machine generates a stream of profits for a firm. Those housing services would not be there otherwise. So, a new house represents an addition to the stock of capital assets and the services they will provide. (Purchase of a second-hand house doesn’t count as new capital for this same reason. Buying a second-hand house just transfers ownership of the services that were already there. No new production and therefore no additional capital production is generated by such a purchase.)

Of course, if it is decided that buying a new house counts as investment, you might ask, what about buying other goods that will provide a stream of services in the future. The problem is, where to stop? Buying a new washing machine provides a stream of “washing services” in the future, but that’s also counted as consumption and not investment. What about a new pen? Doesn’t that provide “writing services” in the future?

As you can see, all sorts of little ambiguities arise once we start defining consumer capital. However, economists have to draw a line somewhere. For whatever reason — perhaps because housing is such an economically important service — buying a new house is recorded as investment, whereas almost anything else a household would buy is considered consumption.

Investment spending and the interest rate

Like all the components of aggregate demand, investment spending is a flow. In the case of investment I, it’s the flow or speed per year with which we are adding new capital to the existing capital stock. We know from Chapter 6 that the ideal capital stock is one large enough to drive capital’s marginal product MP_K to the point where it just covers the real interest rate r plus depreciation δ. That is,

is a necessary condition to be at the point where the capital input is profit maximizing.

Because capital is subject to diminishing returns, a fall in r on the right-hand side of this equation will raise the desired or ideal capital stock K* because that’s what it takes to drive the left-hand-side down as well. So, we can write that the desired capital stock K* is a (negative) function of the interest rate. That is, K* = K*(r). As r rises, K* falls.

Each firm and therefore the macro economy overall enters the current period, with the capital stock that it inherited from last period, K_t-1. So, the gap between the capital firms want to have and what they have is K*(r) – K_t-1. Of course, it’s unlikely that businesses will want to close this gap all in one period. Installing and integrating new plant and equipment with the old stuff is costly and gets more so the quicker you do it. As in Chapter 6, we imagine that firms will want to close only some fraction λ of the gap, where λ might be 0.3 or 0.5 or some other value less than 1. Firms will also need to invest to replace the part of inherited capital being lost to depreciation δK_t–1. Putting this all together, we get the investment equation discussed in Chapter 6:

In this investment equation, the capital stock K_t–1 as of the end of last period is given by history. Additionally, both λ and δ are given parameters. So, the only thing that’s really variable on the right-hand side is the desired capital term, which depends on the real interest rate r. As r falls (rises), K* rises (falls), and the gap between the desired and actual capital stock grows (shrinks). Hence, the investment done this period to close the gap I_t goes up as the real interest rate r goes down and down as r goes up. Because all the other variables are given, we can simplify this by just writing a very parsimonious short-run investment function:

This is just a mathy way of saying that the amount of investment spending in the short run depends primarily on the real interest rate r. In fact, it’s a negative function of the interest rate, because a fall in the interest rate leads to an increase in investment.

Imagine for example that you work for a firm, and you have a great idea for a new investment project. You carefully draw up a proposal and pitch it to your boss. One of the first questions she asks is “How much will it cost?”, closely followed by “How much return will it make the company?”. You reply that it’ll cost $1 million and that you estimate that if the project goes well the firm will see a 5 percent return in one year. Your boss replies that interest rates are 6 percent and the company would be better off just putting that money in the bank. But if interest rates are only 2 percent, your investment project suddenly becomes attractive.

Figure 10-2 displays a linear form of the investment function graphically. Notice that as the interest rate falls, more investment projects start to look attractive and firms purchase more capital stock.

Residential investment by households is also very sensitive to the interest rate. Typically, people need to take out a mortgage in order to purchase a new home. If the interest rate is high, new houses become unaffordable to many people.

Other determinants of investment

Other factors can affect the level of investment in an economy besides the interest rate. In the longer run, institutional features like having secure property rights and an efficient legal system are critical. Whatever Karl Marx may have wished, capital is property, and firms will only invest in that property if they believe their rights to the use of those capital goods and the claims to any income they generate are secure. Likewise, an efficient legal system in which any contract under which the capital is employed can be enforced at relatively low cost is also very important.

In the short run, two other factors loom large. One is taxes, specifically the taxes on capital income such as corporate taxes. If firms know that, say, 50 percent of any profit they make from a new machine is going to be paid as tax, they’re going to be less likely to buy it. Formally, if τ_K is the tax on capital income, then our condition for the ideal capital stock becomes this:

From this, it’s easy to see that as τ_K goes up, the after-tax marginal product of capital goes down. Thus, the desired capital stock must fall to a level where capital’s marginal product is high enough that it can afford to pay the tax and still cover the real rate plus depreciation.

The other factor that influences investment in the short run is expected output itself. If businesspeople expect a strong economy with lots of economic activity, then it makes sense to invest in capital machinery to help meet those anticipated demands. Conversely, in a recession where real GDP is low, and some plants are already sitting idle, it makes little sense to invest in new plant and equipment. You can barely use the capital you have.

Factoring in government purchases

Government expenditure G is the second-largest component of aggregate demand, ranging from 18 to 20 percent of GDP in the U.S. It refers to the government purchasing goods and services. The idea is that these goods and services are eventually provided to households in some shape or form. Examples of government purchases are spending on healthcare, education, defense, roads, and so on.

Social Security spending on benefits and welfare payments aren’t counted as part of government purchases. This is because they’re transfer payments from one group of people to another group: They don’t involve the government purchasing any goods or services.

Government expenditure is usually exogenous in macroeconomic models, which means that the model doesn’t attempt to explain it. It just equals what it equals. This makes sense, because the level of government spending is a political decision taken by politicians – within certain constraints, of course. For this reason G is assumed to be fixed, unless it’s specifically changed by policy-makers.

To show that a variable is fixed, you can put a “bar” above it:

Of course, G does change over time depending on the particular preferences and constraints of the government at the time. But unless you’re explicitly told that government spending has changed, you can assume that it’s constant and unaffected by other macroeconomic variables.

Coming in, going out: Net exports

Net exports NX are exports X minus imports M. This is the smallest of the four components of aggregate demand. For the U.S., it’s generally a small negative number, averaging about 3 percent of GDP over the last 10 years. This doesn’t mean foreign trade is insignificant. Admittedly, its contribution to aggregate demand is relatively small. Even so, there’s a lot of trading going on behind that net figure. Over the past 10 years, exports have averaged 13 percent of GDP while imports have averaged 16 percent. For the many countries that are much smaller than the U.S., the influence of the trading sector is even more considerable.

The main macroeconomic variable that affects net exports is the real exchange rate R. We talk about this variable in Chapter 9 without really naming it. Basically, R tells you the relative price of foreign and domestic goods. It’s distinct from the nominal exchange rate E, which tells you the domestic currency price of foreign currency. For example, if it takes $1.50 to buy £1, we say E = $1.50/£1.

As in Chapter 9, we denote the prices of U.S. and U.K. goods as P^US and P^UK, respectively. The real exchange rate R is then:

We know from Chapter 9 that in the long run, purchasing power parity (PPP) holds apart from any significant trade barriers. So, assuming U.S. and U.K. goods are close substitutes, R should equal 1. In other words, the nominal exchange E should offset any price level differences, implying that E = P^U.S./P^UK in the long run. This For Dummies book should sell for the same dollar price whether bought in the U.S. or bought in the U.K. after converting pounds to dollars.

PPP does not hold that tightly in the short run, though, largely because domestic and foreign price levels such as P^U.S. and P^UK are “sticky.” Consider again the example from Chapter 9 with E = $1.50/1£ and the good in question a For Dummies book that sells for $30 in the U.S. and £20 in the U.K. That’s consistent with a long-run equilibrium because it implies R = 1. If anyone sold a copy in Britain for £20, they could convert into $30 and buy an exact duplicate of the book in the U.S. with no gain or loss.

But suppose the dollar suddenly depreciates, that is, E rises, perhaps due to currency speculation in the foreign exchange markets. It now takes more dollars to buy a pound. To make the example more concrete, we’ll assume that the depreciation is 10 percent or 15 cents to $1.65/£1. If the goods prices don’t change, that is, if the book’s prices stay at $30 and £20, respectively, then there’s a real incentive for Brits to buy the book in the U.S. Instead of spending £20 in the U.K., they can convert their £20 into $33, buy the text for $30, and pocket $3 (or £1.82) at the end of the day. Likewise, any U.S. citizen who was thinking about buying the text in the U.K. will now find that it will take $33 to get the necessary pounds, which is $3 more than the book costs in the U.S.

More generally, as the domestic currency depreciates (E rises), it becomes relatively cheaper for people in the U.K. to buy U.S. goods and more expensive for U.S. citizens to buy U.K. goods. This causes an increase in U.S. exports and a fall in U.S. imports. Hence, net exports NX rise. (Strictly speaking, something called the Marshall-Lerner condition needs to hold for a depreciation to increase net exports. In this book we always assume that the condition holds. (See the nearby sidebar “Delving into the Marshall-Lerner condition” for more details.)

Mathematically, you can express the relationship between net exports and the real exchange rate as follows:

In words, this expression says that net exports (NX) are a function of the real exchange rate R.

Figure 10-3 illustrates the NX(R) relation graphically. You can see that, as R depreciates, net exports increase. The dashed line at 0 (zero) represents the case of balanced trade, that is, imports are exactly matched by exports. To the right of the dashed line, net exports are positive and the economy has a trade surplus: Exports are greater than imports. To the left of the dashed line, net exports are negative and the economy has a trade deficit: Imports are greater than exports. A key point to remember is that in the short run, domestic and foreign goods prices, P^U.S. and P^UK, are sticky. So, movements in the nominal exchange rate E translate quickly into movements in the real exchange rate R.

Observing movement along the AD curve

A movement along the AD curve represents the change in aggregate demand caused only by a change in the price level. We’ve not said where that comes from yet but, if for some reason it happens, that’s what we’re seeing as we move along the AD curve. For that thought experiment, you assume that everything else remains as it was (this is the ceteris paribus assumption, which is Latin for “all other things equal”).

Figure 10-5 shows that at the price level P₁, aggregate demand is equal to Y₁. If the price level falls to P₂ – so that on average things are cheaper – aggregate demand increases from Y₁ to Y₂ due to the three effects outlined in the preceding section. Assuming that everything else remains as it was means that this increase in aggregate demand can be attributed just to the fall in the price level (and the direct effect this fall has on other variables in the economy). In the figure, at price level P₁, aggregate demand is equal to Y₁. At price level P₂, aggregate demand is equal to Y₂.

Similarly, if the price level were to rise from P₂ to P₁, the AD curve tells you that aggregate demand will fall from Y₂ to Y₁.

Tracing a shift of the curve

Anything that causes aggregate demand to change, other than the price level, leads to a shift of the AD curve. A shift changes the amount everyone wants to buy even if the price doesn’t change.

In Figure 10-6, aggregate demand has increased from AD₁ to AD₂, which means that at any given price level, aggregate demand is now greater. For example, at the price level P₁, aggregate demand used to be equal to Y₁. After the shift of the AD curve, aggregate demand would equal Y₂ if the price level were to remain at P₁. In Figure 10-6, the shift is a parallel one. So, a price change causes the same move along the curve after the shift as before it. That’s common, but not always the case.

Many things can cause the AD curve to shift in this way, including the following:

• Increased consumer confidence: At any given price level, households want to consume more (an increase in C).
• Increased confidence of firms: At any given price level, firms want to invest more (an increase in I).
• Increase in government purchases: Perhaps the government decides that it wants to spend more on infrastructure (an increase in G).
• Increase in demand for exports: Caused by a fall in the value of the dollar (an increase in NX).
• Lower taxes: Encourages spending by households and firms (an increase in C and I).
• Increase in the money supply reduces the interest rate: Stimulates spending by households and firms (an increase in C and I).

Clearly, all these factors also work in reverse as well. That is, a fall in consumer confidence or a fall in government expenditure leads to a decrease in aggregate demand, which shifts the AD curve to the left (see Figure 10-7).