Between 1929 and 1933, there was a sharp fall in aggregate demand—a reduction in the quantity of goods and services demanded at any given price level. One consequence of the economy-wide decline in demand was a fall in the prices of most goods and services. By 1933, the GDP deflator (one of the price indexes we defined in Chapter 7) was 18% below its 1929 level, and other indexes were down by similar amounts. A second consequence was a decline in the output of most goods and services: by 1933, real GDP was 30% below its 1929 level. A third consequence, closely tied to the fall in real GDP, was a surge in the unemployment rate from 2.93% to 19.3%.

The association between the plunge in real GDP and the plunge in prices wasn’t an accident. Between 1929 and 1933, the Canadian economy was moving down its aggregate supply curve, which shows the relationship between the economy’s aggregate price level (the overall price level of final goods and services in the economy) and the total quantity of final goods and services, or aggregate output, producers are willing to supply. (As you will recall, we use real GDP to measure aggregate output. So we’ll often use the two terms interchangeably.) More specifically, between 1929 and 1933, as a result of a leftward shift of aggregate demand, the economy moved down its (upward-sloping) short-run aggregate supply curve.

The period from 1929 to 1933 demonstrated that there is a positive relationship in the short run between the aggregate price level and the quantity of aggregate output supplied. That is, a rise in the aggregate price level is associated with a rise in the quantity of aggregate output supplied, other things equal; a fall in the aggregate price level is associated with a fall in the quantity of aggregate output supplied, other things equal. To understand why this positive relationship exists, consider the most basic question facing a producer: is producing a unit of output profitable or not? Let’s define profit per unit:

(12-2) Profit per unit of output = Price per unit of output − Production cost per unit of output

Thus, the answer to the question depends on whether the price the producer receives for a unit of output is greater or less than the cost of producing that unit of output. At any given point in time, many of the costs producers face are fixed per unit of output and can’t be changed for an extended period of time. Typically, the largest source of inflexible production cost is the wages paid to workers. Wages here refers to all forms of worker compensation, such as employer-paid health care and retirement benefits in addition to earnings.

Wages are typically an inflexible production cost because the dollar amount of any given wage paid, called the nominal wage, in the short run, is often determined by contracts that were signed some time ago. And even when there are no formal contracts, there are often informal agreements between management and workers, making companies reluctant to change wages in response to economic conditions. For example, companies usually will not reduce wages during poor economic times—unless the downturn has been particularly long and severe—for fear of generating worker resentment. Correspondingly, they typically won’t raise wages during better economic times—until they are at risk of losing workers to competitors—because they don’t want to encourage workers to routinely demand higher wages.

As a result of both formal and informal agreements, then, the economy is characterized by sticky wages: nominal wages that are slow to fall even in the face of high unemployment and slow to rise even in the face of labour shortages. It’s important to note, however, that nominal wages cannot be sticky forever: ultimately, formal contracts and informal agreements will be renegotiated to take into account changed economic circumstances. As the Pitfalls at the end of this section explains, how long it takes for nominal wages to become flexible is an integral component of what distinguishes the short run from the long run.

To understand how the fact that many costs are fixed in nominal terms gives rise to an upward-sloping short-run aggregate supply curve, it’s helpful to know that prices are set somewhat differently in different kinds of markets. In perfectly competitive markets, producers take prices as given; in imperfectly competitive markets, producers have some ability to choose the prices they charge. In both kinds of markets, there is a short-run positive relationship between prices and output, but for slightly different reasons.

Let’s start with the behaviour of producers in perfectly competitive markets; remember, they take the price as given. Imagine that, for some reason, the aggregate price level falls, which means that the price received by the typical producer of a final good or service falls. Because many production costs are fixed in the short run, production cost per unit of output doesn’t fall by the same proportion as the fall in the price of output. So the profit per unit of output declines, inducing perfectly competitive producers to look for ways to alter their production plans so as to limit this decline. Since land, capital, and technology costs cannot be lowered, because they are fixed in the short run, the only way to reduce cost is to reduce the cost of labour. And if wages are fixed, this can be done only by reducing the (profit-maximizing) level of labour employed—that is, by laying off workers. The law of diminishing marginal productivity implies that at some point the marginal product of labour declines as the number of workers employed rises. This implies that layoffs raise the marginal productivity of labour, and this higher productivity works to lower per unit production costs.

But layoffs also reduce the level of output, causing a lower short-run revenue, so why do producers do this? It is because they want to maximize profits, regardless of whether they are operating for the short or long term. So when the aggregate price level falls, reducing the profit per unit, producers want to reduce their per unit production costs because that will push their profit per unit back up, somewhat. In the short run, profit-maximizing producers in a perfectly competitive industry operate where their short-run marginal cost curve is at or above their short-run average variable cost curve. So, at their current production level, their short-run average cost curve slopes upward as the level of output is increased. Consequently, a reduction in level of output will lower the per unit production cost, which will help mitigate the drop in the profit per unit.

On the other hand, suppose that for some reason the aggregate price level rises. As a result, the typical producer receives a higher price for its final good or service. Again, many production costs are fixed in the short run, so production cost per unit of output doesn’t rise by the same proportion as the rise in the price of a unit. And since the typical perfectly competitive producer takes the price as given, profit per unit of output rises and producers respond by either hiring workers or increasing the number of hours worked by the existing workers (i.e., more overtime), thus raising the level of output supplied.

Now consider an imperfectly competitive producer that is able to set its own price. If there is a rise in the demand for this producer’s product, it will be able to sell more at any given price. Given stronger demand for its products, it will probably choose to increase its prices as well as its output, as a way of increasing profit per unit of output. In fact, industry analysts often talk about variations in an industry’s “pricing power”: when demand is strong, firms with pricing power are able to raise prices—and they do.

Conversely, if there is a fall in demand, firms will normally try to limit the fall in their sales by cutting prices.

Both the responses of firms in perfectly competitive industries and those of firms in imperfectly competitive industries lead to an upward-sloping relationship between aggregate output and the aggregate price level. The positive relationship between the aggregate price level and the quantity of aggregate output producers are willing to supply during the time period when many production costs, particularly nominal wages, can be taken as fixed is illustrated by the short-run aggregate supply curve. The positive relationship between the aggregate price level and aggregate output in the short run gives the short-run aggregate supply curve its upward slope.

Figure 12-5 shows a hypothetical short-run aggregate supply curve, SRAS, which matches actual Canadian data for 1929 and 1933. On the horizontal axis is aggregate output (or, equivalently, real GDP)—the total quantity of final goods and services supplied in the economy—measured in 2002 dollars. On the vertical axis is the aggregate price level as measured by the GDP deflator, with the value for the year 2002 equal to 100. In 1929, the aggregate price level was 8.1 and real GDP was $78.3 billion. In 1933, the aggregate price level was 6.6 and real GDP was only $54.7 billion. The movement down the SRAS curve corresponds to the deflation and fall in aggregate output experienced over those years.

Figure 12-5 shows a movement along the short-run aggregate supply curve, as the aggregate price level and aggregate output fell from 1929 to 1933 (perhaps caused by an adverse AD shock). But there can also be shifts of the short-run aggregate supply curve, as shown in Figure 12-6. Panel (a) shows a decrease in short-run aggregate supply—a leftward shift of the short-run aggregate supply curve. Aggregate supply decreases when producers reduce the quantity of aggregate output they are willing to supply at any given aggregate price level. Panel (b) shows an increase in short-run aggregate supply—a rightward shift of the short-run aggregate supply curve. Aggregate supply increases when producers increase the quantity of aggregate output they are willing to supply at any given aggregate price level.

To understand why the short-run aggregate supply curve can shift, it’s important to recall that producers make output decisions based on their profit per unit of output. The short-run aggregate supply curve illustrates the relationship between the aggregate price level and the profit-maximizing level of aggregate output supplied: because some production costs are fixed in the short run, a change in the aggregate price level leads to a change in producers’ profit per unit of output and, in turn, leads to a change in the profit-maximizing level of aggregate output supplied. But other factors besides the aggregate price level can affect profit per unit and, in turn, the profit-maximizing level of aggregate output supplied. It is changes in these other factors that will shift the short-run aggregate supply curve.

To develop some intuition, suppose that something happens that raises production costs—say, an increase in the price of oil. At any given price of output, a producer now earns a smaller profit per unit of output. As a result, producers reduce the quantity supplied at any given aggregate price level, and the short-run aggregate supply curve shifts to the left. If, in contrast, something happens that lowers production costs—say, a fall in the nominal wage—a producer now earns a higher profit per unit of output at any given price of output. This leads producers to increase the quantity of aggregate output supplied at any given aggregate price level, and the short-run aggregate supply curve shifts to the right.

Now we’ll discuss some of the important factors that affect producers’ profit per unit and so can lead to shifts of the short-run aggregate supply curve.

Changes in Commodity Prices In this chapter’s opening story, we described how a surge in the price of oil caused problems for the Canadian economy in the 1970s, early in 2008, and again in 2011. Oil is a commodity, a standardized input bought and sold in bulk quantities. An increase in the price of a commodity—oil— raised production costs across the economy and reduced the quantity of aggregate output supplied at any given aggregate price level, shifting the short-run aggregate supply curve to the left. Conversely, a decline in commodity prices reduces production costs, leading to an increase in the quantity supplied at any given aggregate price level and a rightward shift of the short-run aggregate supply curve.

Why isn’t the influence of commodity prices already captured by the short-run aggregate supply curve? Because commodities—unlike, say, soft drinks—are not a final good, their prices are not included in the calculation of the aggregate price level. Further, some commodities are a key raw material input and thus represent a significant cost of production to some goods suppliers, just like nominal wages do. So changes in certain commodity prices can have large impacts on production costs. And in contrast to non-commodities, the prices of commodities can sometimes change drastically due to industry-specific shocks to supply—such as wars in the Middle East or rising Chinese demand that leaves less oil, iron, nickel ores, potash, and so on, for Canada and the rest of the world.

Changes in Nominal Wages At any given point in time, the dollar wages of many workers are fixed because they are set by contracts or informal agreements made in the past. Nominal wages can change, however, once enough time has passed for contracts and informal agreements to be renegotiated. Suppose, for example, that there is an economy-wide rise in the cost of health care insurance premiums paid by employers as part of employees’ wages. From the employers’ perspective, this is equivalent to a rise in nominal wages because it is an increase in employer-paid compensation. So this rise in nominal wages increases production costs and shifts the short-run aggregate supply curve to the left. Conversely, suppose there is an economy-wide fall in the cost of such premiums. This is equivalent to a fall in nominal wages from the point of view of employers; it reduces production costs and shifts the short-run aggregate supply curve to the right.

An important historical fact is that during the 1970s the surge in the price of oil had the indirect effect of also raising nominal wages. This “knock-on” effect occurred because many wage contracts included cost-of-living allowances that automatically raised the nominal wage when consumer prices increased. Through this channel, the surge in the price of oil—which led to an increase in overall consumer prices—ultimately caused a rise in nominal wages. So the economy, in the end, experienced two leftward shifts of the aggregate supply curve: the first generated by the initial surge in the price of oil, the second generated by the induced increase in nominal wages. The negative effect on the economy of rising oil prices was greatly magnified through the cost-of-living allowances in wage contracts. Today, automatic cost-of-living allowances in wage contracts are rare.

Changes in Productivity An increase in productivity means that a worker can produce more units of output with the same quantity of inputs. For example, the introduction of bar-code scanners in retail stores greatly increased the ability of a single worker to stock, inventory, and resupply store shelves. As a result, the cost to a store of “producing” a dollar of sales fell and profit rose. And, correspondingly, the quantity supplied increased. (Think of Walmart and dollar stores, such as Dollarama, and the increase in the number of their stores as an increase in aggregate supply.) So a rise in productivity, whatever the source, increases producers’ profits and shifts the short-run aggregate supply curve to the right. Conversely, a fall in productivity—say, due to new regulations that require workers to spend more time filling out forms—reduces the number of units of output a worker can produce with the same quantity of inputs. Consequently, the cost per unit of output rises, profit falls, and quantity supplied falls. This shifts the short-run aggregate supply curve to the left.

For a summary of the factors that shift the short-run aggregate supply curve, see Table 12-2.

We’ve just seen that in the short run a fall in the aggregate price level leads to a decline in the quantity of aggregate output supplied because nominal wages are sticky in the short run. But, as we mentioned earlier, contracts and informal agreements are renegotiated in the long run. So in the long run, nominal wages— like the aggregate price level—are perfectly, or fully, flexible, not sticky. This fact greatly alters the long-run relationship between the aggregate price level and aggregate supply. In fact, in the long run the aggregate price level has no effect on the quantity of aggregate real output supplied.

To see why, let’s conduct a thought experiment. Imagine that you could wave a magic wand—or maybe a magic bar-code scanner—and cut all prices in the economy in half at the same time. By “all prices” we mean the prices of all inputs, including nominal wages, as well as the prices of final goods and services. What would happen to aggregate real output, given that the aggregate price level has been halved and all input prices, including nominal wages, have been halved?

The answer is: nothing. Consider Equation 12-2 again: each producer would receive a lower price for its product, but costs would fall by the same proportion. As a result, every unit of output profitable to produce before the change in prices would still be profitable to produce after the change in prices. So a halving of all prices in the economy has no effect on the economy’s aggregate real level of output. In other words, in the long-run, changes in the aggregate price level have no effect on the quantity of aggregate output supplied.

In reality, of course, no one can change all prices by the same proportion at the same time. But now, we’ll consider the long run, the period of time over which all prices are perfectly, or fully, flexible. In the long run, inflation or deflation has the same effect as someone changing all prices by the same proportion. As a result, changes in the aggregate price level do not change the real quantity of aggregate output supplied in the long run. That’s because changes in the aggregate price level will, in the long run, be accompanied by equal proportional changes in all input prices, including nominal wages.

The long-run aggregate supply curve, illustrated in Figure 12-7 by the LRAS curve, shows the relationship between the aggregate price level and the quantity of aggregate output supplied that would exist if all prices, including nominal wages, were perfectly flexible. The long-run aggregate supply curve is vertical because changes in the aggregate price level have no effect on aggregate output in the long run. At an aggregate price level of 9.0, the quantity of aggregate output supplied is $80 billion in 2002 dollars. If the aggregate price level falls by 50% to 4.5, the quantity of aggregate output supplied is unchanged in the long run at $80 billion in 2002 dollars.

It’s important to understand not only that the LRAS curve is vertical but also that its position along the horizontal axis represents a significant measure. The horizontal intercept in Figure 12-7, where LRAS touches the horizontal axis ($80 billion in 2002 dollars), is the economy’s potential output, Y_P: the level of real GDP the economy would produce if all prices, including nominal wages, were perfectly, or fully, flexible.¹

In reality, the actual level of real GDP is almost always either above or below potential output. We’ll see why later in this chapter, when we discuss the AD-AS model. Still, an economy’s potential output is an important number because it defines the trend around which actual aggregate output fluctuates from year to year.

In Canada, the office of the Parliamentary Budget Officer, or PBO, estimates annual potential output for the purpose of federal budget analysis.² In Figure 12-8, the PBO’s estimates of Canadian potential output from 1980 to 2011 are represented by the orange line and the actual values of Canadian real GDP over the same period are represented by the blue line. Years shaded purple on the horizontal axis correspond to periods in which actual aggregate output fell short of potential output, while years shaded green correspond to periods in which actual aggregate output exceeded potential output.

The PBO is not the only entity to forecast Canada’s potential GDP: the Bank of Canada, the Federal Department of Finance, the OECD, the International Monetary Fund (IMF), and many private sector forecasters, such as the Royal Bank, do likewise. These forecasts are used for several purposes, including the measurement of the output gap. The output gap is the actual real (measured) GDP minus potential GDP, often calculated and presented as a percentage of potential GDP.

The output gap helps agents gauge the performance of the economy relative to the long-run (potential) trend. When the output gap is negative, actual real output is below the long-run potential level, so the economy is underperforming relative to the long-run trend. Similarly, when the output gap is positive, actual real output has risen above its long-run potential level, and the economy is booming relative to the long-run trend or potential level of GDP. A positive output gap implies greater inflationary pressure exists in the economy—the larger the gap, the more inflationary pressure is implied.

In 2010, the PBO used its estimates of potential GDP and the associated output gap along with similar estimates calculated by the IMF to conclude that the severity of the 2008-2009 recession in Canada was in line with the experiences of other G-7 countries. That is, the recession did cause our output gap to become negative as actual GDP fell below potential GDP, but we fared no worse than other G-7 countries with respect to the size of our output gap in terms of the percentage and duration of the reduction. On the other hand, the recession hit Japan harder than the rest of the G-7: its output gap fell to almost -7% of potential GDP, about two full percentage points lower than the gap for Canada, the U.K., and the U.S., but Japan’s economy was already weaker before the recession.

As you can see, Canadian potential output has risen steadily over time— implying a series of rightward shifts of the LRAS curve. What has caused these rightward shifts? The answer lies in the factors related to long-run growth that we discussed in Chapter 9, such as increases in physical capital and human capital, the size of the labour force, and technological progress. Over the very long run, as the amount of both physical and human capital, size of the labour force, and total factor productivity all rise, the level of real GDP that the economy is capable of producing also rises and this causes a rightward shift of the LRAS curve. Figure 12-9 shows a rightward shift in the LRAS curve. Indeed, one way to think about long-run economic growth is that it is the growth in the economy’s potential output. We generally think of the long-run aggregate supply curve as shifting to the right over time as an economy experiences long-run growth.

For a summary of the factors that shift the long-run aggregate supply curve, see Table 12-3.

As you can see in Figure 12-8, the economy normally produces more or less than potential output: actual aggregate output was significantly below potential output in the recessions of the early 1980s and the early 1990s, significantly above potential output in the boom of the late 1980s and in 2000, and significantly below potential output after the recession of 2008-2009. So the economy is normally on its short-run aggregate supply curve—but not on its long-run aggregate supply curve. So why is the long-run curve relevant? Does the economy ever move from the short run to the long run? And if so, how?

The first step to answering these questions is to understand that the economy is always in one of only two states with respect to the short-run and long-run aggregate supply curves. It can be on both curves simultaneously by being at a point where the curves cross (as in the few years in Figure 12-8 in which actual aggregate output and potential output roughly coincided). Or it can be on the short-run aggregate supply curve but not the long-run aggregate supply curve (as in the years in which actual aggregate output and potential output did not coincide). But that is not the end of the story. If the economy is on the short-run but not the long-run aggregate supply curve, the short-run aggregate supply curve will shift over time until the economy is at a point where both curves cross—a point where actual aggregate output is equal to potential output.

Figure 12-10 illustrates how this process works. In both panels LRAS is the long-run aggregate supply curve, SRAS₁ is the initial short-run aggregate supply curve, and the aggregate price level is at P₁. In panel (a) the economy starts at the initial production point, A₁, which corresponds to a quantity of aggregate output supplied, Y₁, that is higher than potential output, Y_P. Producing an aggregate output level (such as Y₁) that is higher than potential output (Y_P) is possible only because nominal wages haven’t yet fully adjusted upward. Until this upward adjustment in nominal wages occurs, producers are earning high profits and producing a high level of output. But a level of aggregate output higher than potential output means a low level of unemployment. Because jobs are abundant and workers are scarce, nominal wages will rise over time, gradually shifting the short-run aggregate supply curve leftward. Eventually it will be in a new position, such as SRAS₂. (Later in this chapter, we’ll show where the short-run aggregate supply curve ends up. As we’ll see, that depends on the aggregate demand curve as well.)

In panel (b), the initial production point, A₁, corresponds to an aggregate output level, Y₁, that is lower than potential output, Y_P. Producing an aggregate output level (such as Y₁) that is lower than potential output (Y_P) is possible only because nominal wages haven’t yet fully adjusted downward. Until this downward adjustment occurs, producers are earning low (or negative) profits and producing a low level of output. An aggregate output level lower than potential output means high unemployment. Because workers are abundant and jobs are scarce, nominal wages will fall over time, shifting the short-run aggregate supply curve gradually to the right. Eventually it will be in a new position, such as SRAS₂.

We’ll see shortly that these shifts of the short-run aggregate supply curve will return the economy to potential output in the long run.