Aggregate supply is the aggregate of all the supply in the economy. Hence, the aggregate supply (from now on, AS) curve is the sum of all the industry supply curves. It shows the relationship between the price level and real output (or real national income).
When we looked at firm and industry cost curves (see the 'Costs and revenues' topic and the relevant 'Market structure' topic) it was important to distinguish between the short and long run. We have to do this again here, in particular, the prices of all factor inputs (including wage rates) are assumed to be constant in the short run.
It makes sense, therefore, that in the short run the AS curve will be upward sloping, just like the industry supply curves. The obvious reason is that all the industry supply curves that are added together are upward sloping too. There is a more technical explanation, though.
If real output is to increase in the short run, firms cannot attract new labour by increasing the wage rate. The only way to make more output is to get the current workforce to work harder (overtime, perhaps). In the real world, overtime pays double the wage rate, or maybe time and a half. So the costs to the firms and industries are likely to rise as real output rises. These increased costs will tend to be passed onto the consumer through higher prices. In a roundabout way, the increase in real output has resulted in a rise in the price level.
The short run AS curve shifts for similar reasons as a firm's, or an industry's supply curve might shift. Anything that causes the costs of the industries within the economy to change (regardless of changes in the price level) will shift the AS curve. The factors must have nothing to do with changes in the general price level, remember. Otherwise, we would be dealing with a movement along the AS curve and not a shift in the curve.
In the diagram you can see that the short run AS curve (or SRAS curve) has shifted upwards (or to the left) from SRAS1 to SRAS2. There are two different ways that this can be interpreted.
You could argue that, for a given level of real output, Y1, the price level has risen due to a rise in the costs of various industries (like an increase in wage rates, other input prices or even the tax imposed by the government) from P1 to P2.
Another way of looking at it is that real output falls for a given price level due to the increases in industries' costs. The firms may be unable to push the cost increases onto the consumer (for whatever reason) so real output cannot remain at Y1, and so it falls to Y2. The economy cannot make as much real output at the given price level if their costs rise.
Just like with AD curves, the SRAS curves can also shift to the right. Obviously, this would happen if there were an increase in costs for firms in the economy as a whole.
Given that we are assuming that factor input price should remain constant in the short run, the reasons given for the shifts in the SRAS curve do not make much sense. In the real world, industries and economies are often on the receiving end of supply side shocks. There is nothing they can do about them and they affect the costs of the industries or economies. The quadrupling of the price of oil in the 70s is a good example. The industries of the UK could do nothing about it, but it increased the costs of nearly all firms in all industries at the time, causing the SRAS curve to shift to the left.
Different economists have different views about the long run AS curve (LRAS curve). You have probably come across the differences between Monetarist (or classical) economists and Keynesian economists in your studies. The two sets of economists have different views about the shape of the LRAS curve.
The classical (or monetarist) view
Classical economists believe strongly in the concepts of supply and demand in all markets. Whilst they are prepared to believe that markets might not be in equilibrium in the short run, they believe that all markets clear (i.e. supply equals demand) in the long run. They also believe that if all markets are working efficiently and all resources are being used, then the economy must be at a point on their production possibility frontier (PPF). In other words, the economy is making as much as it can, given its resources.
Classical economists think that the LRAS curve is vertical. There is absolutely no spare capacity in the economy. The maximum real output is already being produced. In these circumstances, if there is an increase in AD (a shift from AD1 to AD2) for whatever reasons then all that will happen is the price level will rise (from P1 to P2). Demand increases, but there is no more output to buy, so the price of the various goods and services is bid up.
Note that the economy is in equilibrium where the AD curve cuts the AS curve (in this case, the LRAS curve). This occurs at point A initially, and then at point B after the shift in the AD curve from AD1 to AD2.
The Keynesian view
John Maynard Keynes (1883 - 1946) is probably the most famous economist of the twentieth century. Until the depression of the 1930s, all economists were effectively classical. Unemployment rose to astronomical levels during the depression (a very bad recession!). The classical economists, with their belief in supply and demand up their sleeves, said that the labour market would 'clear'. Unemployment was caused by excess supply of labour due to the wage rate (the price) being too high. Before long, the wage rate would fall as workers realised that the only way to get work was to take a lower wage. Of course, when this didn't happen after a number of years, they could not explain why.
Keynes was an economist who was always looking for answers to problems. He didn't care that every other economist in the world said that wages would fall and, with them, unemployment. As far as he was concerned, it wasn't happening, so the theory was wrong. He argued that an economy could find itself in a situation where unemployment was high and stuck at the high rate. In other words, the economy could find itself in an equilibrium (i.e. a state of rest) that was not the full employment equilibrium.
In the diagram above, the levels of real output between 0 and Y1 are ones where there is a lot of excess capacity. There are many unused resources (unemployed labour, for example) and the economy is at a point well within its PPF (see the topic called 'Market failure' for details of the PPF). This was the situation in the depression of the 30s. Keynes' solution was simple. If there is excess supply, the solution is not to hope the wage (and other prices) fall, causing the supply of labour to be offered onto the market to fall. Why not close the gap by increasing demand? Consumers were too poor at the time to create this increase in AD, so, Keynes argued, it was up to the government to create this demand, through investment in roads, railways, hospitals, etc. Not only would this create jobs for the unemployed, but it would also improve the infrastructure of the economy.
Anyway, diagrammatically, you can see that, up to point A, the AD curve can shift to the right without the price level rising. A large increase in real output with no rise in the price level. Also note that at every point where an AD curve cuts the LRAS curve is an equilibrium point. Importantly (as Keynes said), these equilibrium levels of real output can occur at levels below the full employment level of real output (like the equilibrium at point C, price P1 and real output Y1).
The section of the LRAS curve between points A and B shows the levels of output where the economy is coming out of recession. If AD continues to rise (a shift to AD2, say), real output will still rise, but the price level will also rise. In this situation, some industries will still be experiencing excess capacity, but others will be finding that they are getting close to full capacity. So some prices will not be rising, but others will be. The price level, remember, is a measure of the average price level across the whole economy. So, in this situation, the price level will rise.
The final section of the LRAS curve, above point B, is vertical. Keynes agreed with the classical economists that, once the economy had reached the full employment level of real output, any rise in AD will be inflationary. A shift in AD from AD3 to AD4 will result in no increase in real output, but the price level will rise from P3 to P4. This last bit of the Keynesian LRAS curve is the same as the whole of the classical LRAS curve.
Comparisons with the 45 degree diagram analysis
For those of you who prefer the 45 degree diagram analysis, look at the diagram below and see how it compares with the AS/AD diagram above.
The Y = AD line at 45 degrees to the horizontal gives all the points where the economy is in equilibrium, just like all the points where the various AD curves cut the LRAS curve are equilibrium points in the previous diagram. Notice that each AD line in this diagram denotes planned expenditure, which is C + I + G + X − M.
Note that AD3 gives the full employment level of real national income (YFE), just as the AD curve, AD3, does as is crosses the LRAS curve at point B in the previous diagram.
The AD4 lines are showing the same thing as well. In the first diagram you see the price rise to P4, and in the second diagram this rise in the price level is illustrated by the inflationary gap in green.
The AD1 lines are also showing the same thing. In the first diagram, this gives a level of real output of Y1, which is below the full employment level, and in the second diagram it does the same. The deflationary gap in blue illustrates the downward pressure on prices of being in this recessionary position.
You can probably see why examiners prefer the AS/AD approach. It shows both the Keynesian position and the classical position (when the LRAS curve is vertical), whereas the 45-degree diagram is just a Keynesian diagram. Also, when AD shifts to a position past AD3, real output remains at the full employment level YFE. This doesn't really happen on the 45-degree diagram. The AS/AD diagram also shows the price levels. The 45-degree diagram hints at what is going on with prices with its deflationary and inflationary gap, but it is not specific.