You can use the ASAD to model the effect of a supply shock. The most common and relevant one to look at would be a rise in the price of oil as this is one that affects economies around the world pretty often.
The way to think about this is to think about the supply shock affecting the ‘mark-up’ in firms’ price setting decisions. If the price of oil rises, that will push firms’ costs up outside of labour costs. The mark-up, , is not just a ‘profit’ mark-up, but captures the costs of other inputs as well.
Remember that a negative supply shock, which pushes up the profit mark-up, will cause the AS curve to shift up.
Lets see what’s going on here:
We’ve shifted from curve AS1 to AS2 which has pushed prices up from their original level (where prices were equal to expected prices, Pe), to the new price of P2. Output has fallen from the natural level of output Yn to Y2. So we have had prices going up and output going down in the short run.
It gets worse in the medium run. Remember that the starting medium run equilibrium was where prices equal expected prices. The shift of the AS curve has implied that if the supply shock is permanent, we will go to a new medium run equilibrium where prices equal expected prices at a new natural level of output, Yn’. Obviously if the supply shock is temporary, the AS curve will just shift back, but if it is permanent then firms will always face higher costs, so at any given level of prices, they will be able to produce less. This is what the shift of the AS curve has told us.
So this supply shock has meant higher prices, lower output, and if the shock is permanent, a new lower natural level of output for the economy. Supply shocks are generally bad news.
Now at this point the policy makers have a decision to make. Do they use expansionary fiscal or monetary policy, to try and increase output, or do they just leave things alone and see what happens.
If they don’t do anything, the AS curve will just keep on shifting upwards until it settles at the point where it intersects the AD curve at the new natural level of output, Yn’.
So we end up with a new expected price level, Pe=P3, where price expectations have caught up with prices, in a new medium run equilibrium, at the new natural level of output, Yn’.
Now what happens if the policy makers try to use an expansionary fiscal or monetary policy to counter the loss of output?
In the short run, this expansionary policy managed to limit the loss of output – notice how Y2 is closer to Yn in this diagram than the one above. But the price that they pay for that is that when the economy does settle at the new natural level of output, Yn’, the price level will be even higher than it was on the one above. So whilst the expansionary policy can help limit a collapse in output in the short run, when it comes to the medium run you will pay for it with higher prices. Of course there may be other reasons which motivate using an expansionary policy, you might reason that allowing output to collapse could cause permanent scars to the economy, leave a generation unemployed that will never get back into the economy, drive foreign investment out of the country and so on, which would mean your AS curve actually carried on shifting further to the left. But the basic analysis of the model is that expansionary policy would be a bad thing in terms of prices in the medium run.
What about if the policy makers actually used a contractionary policy?
This time the collapse in output in the short term is much sharper. But the eventual rise in prices by the time you get to the medium run is not as pronounced. So if you don’t mind compounding the problem of falling output in the short run by using a contractionary policy, you can get a lower price rise by the time you get to the medium run. In practice this sort of policy would not be politically very popular, but it works as a theoretical example in the model.
At A-level you might have seen the AS curve presented in the form of a ‘short run’ AS curve which is horizontal, and a ‘long run’ AS curve which is vertical. In Blanchard’s textbook he presents it in a much more logical way – an upward sloping AS curve, which moves up and down according to expected price level. This is basically how you get from a horizontal short-run, to a vertical long-run, except Blanchard defines ‘long-run’ as ‘medium-run’, because long-run really means long term growth models like the Solow growth model that are not concerned with economic fluctuations but with the growth of the long run supply potential of the economy.
The upward sloping AS curve basically shows how output influences prices. As output rises, this drives up prices, which is one explanation for why you generally get inflation in a growing economy.
Now given that the AS model has P on the vertical axis and Y on the horizontal axis, if we are going to draw a curve from that relation, we need Y to be in there somewhere. Well it sort of is, because u is in there, we could express u as a function of Y.
The logic here is that there will be an inverse relation between u and Y. When there is greater demand for goods and Y increases, more workers are hired and so u falls. When Y is falling, there is less need for workers so workers are made redundant, and u rises. So really u in the AS relation kind of represents output, but in an inverse way, as rising Y means falling u.
So how does the AS curve link output to prices?
1 As output increases, more workers are hired to produce the extra output, so unemployment falls.
2 As unemployment falls, workers get more bargaining power, so the WS relation shifts up, wages rise.
3 As wages rise, the PS relation implies that higher W means higher P, so higher wages make prices rise.
4 As workers see prices are rising then expected prices rise. The WS relation implies that higher expected prices mean higher wages, so wages rise.
5 Return to part 3 and repeat.
This is the wage-price spiral and one reason why wages are a large component of driving inflation.
The whole AS curve moves up or down depending on changes in expected prices, the ‘mark-up’ or the ‘catch-all’ variable:
The AS curve will shift up if expected prices rise, or if there is a negative supply shock which raises firms’ costs (eg rise in price of oil) and raises the ‘mark-up’. It would also shift up if there were changes in the catch-all variable that basically meant workers had more bargaining power to negotiate higher wages (eg higher trade union power, higher benefits etc).
The AS curve will shift down if expected prices fall, or if there is a positive supply shock which lowers firms’ costs (eg fall in price of oil, or development of new technology that cuts firms’ costs) and lowers the ‘mark-up’. The mark-up could also be lowered if you moved to a situation where you had more competitive markets. Changes in the catch-all variable that basically meant workers had less bargaining power to negotiate higher wages (eg lower trade union power, reduced benefits etc) would also shift the AS curve down.
The ‘natural’ rate of unemployment is one of those controversial concepts that will immediately provoke angry rants from the anti-capitalists in any economics class when it gets brought up. The answer from the hard core free-marketeers would be that unemployment is a signal of market failure, and that if you had perfectly working labour markets (no minimum wage, no benefits, no trade unions etc) then wages would fall to a low enough level to eliminate unemployment. I will leave this moral debate alone here and just focus on how this labour market model explains the natural rate of unemployment…
For the wage-setting relation, we will have to assume for this. This is often seen as a medium-run assumption. When you have something in economics being ‘expected’, like expected prices, expected inflation, the idea is usually that the expected and the actual values can differ in the short-run, but peoples’ expectations won’t be wrong indefinitely, eventually expectations will adjust to the point where expectations catch up with reality, this is the ‘medium run’.
So in the medium-run the wage-setting relation is , so .
And the price-setting relation is
The natural rate of unemployment is defined as the unemployment rate such that the real wage determined in wage-setting and price-setting relations is the same – this is equilibrium in the labour market. You can represent this with a diagram with (W/P) on the vertical axis and u, unemployment, on the horizontal axis.
The rate at which the two come into equilibrium gives the natural rate of unemployment.
The wage-setting curve is downward sloping, capturing the fact that higher unemployment leads to lower wages. So what about the other variables in this model, the ‘catch-all’ variable and the ‘mark-up’?
If things in the ‘catch all’ variable change, this will shift the wage-setting curve up or down. For instance if there is an increase in trade union power, or increase in benefits, this will increase workers’ bargaining power, so it will shift the WS curve up. This will lead it to intersect the price-setting curve at a higher ‘natural’ rate of unemployment. This is the issue of ‘insiders v outsiders’ that sometimes comes up in relation to debates about trade unions, often when unions get more power, they increase the wages of those already working, at the expense of making it more difficult for outsiders to find work in the industry.
If things in the ‘mark-up’ change (such as an increase in non-labour costs for firms, or an increase in market power allowing them to charge prices further above costs) then this will shift the price-setting relation, but remember that because in terms of real wages, , the mark-up is on the denominator here, so those cases mentioned above which would increase the mark-up, increase the denominator of this equation so they shift the PS curve downwards. This will also increase the natural rate of unemployment.
Firms ultimately want to make economic profits, ie they want their revenues to exceed their costs. This isn’t always possible – if they are in a competitive market and are price takers then they have to accept the price the market dictates. In the real world however, most firms have at least some element of market power because they can differentiate their products from their rivals. If a firm has some market power it has the ability to have some control over the prices it charges.
This gives us the price-setting relation:
This is basically saying that firms charge some form of ‘mark-up’ (denoted by ) over the wages they pay. The wages, which are determined by wage-bargaining, represent part of a firm’s costs, but they are unlikely to be the only costs they face.
So we can think of the ‘mark-up’ as including two elements, it includes the other input costs that firms face apart from labour costs, and also it includes the element of market power that the firm has to charge a price above its marginal cost (ie its Lerner Index).
The mark-up in this model is pretty useful because we can use it to model what will happen if we have other supply shocks, eg a rise in the price of oil, will mean firms have to put the costs up independently of wages (so we represent this with an increase in the mark-up), and we can also capture a change in the market structure which would give a firm more power (forming a cartel, legislation which makes it harder for new firms to enter, an increase in advertising or R&D which makes it harder for new firms to enter etc) – again this can be represented with an increase in the mark-up.
Usually wages are determined by a process of bargaining between workers and firms. The level of wages will depend upon the respective bargaining power of each side, and so conditions in the labour market will come into play here. When unemployment is high, it’s a ‘buyers market’ for firms wanting to hire workers, they know they have a large pool of unemployed labour to choose from so they can offer lower wages. Workers will take the jobs because they are just happy to have jobs. On the other hand, when unemployment is low, firms have to try to compete with each other for workers, so this bids up wages. Workers who aren’t happy with their wage offer at a particular firm will just go and work for a rival, and low unemployment means they might be quite hard to replace.
Blanchard uses this equation as the wage-setting relation: .
This basically means wages are equal to expected prices multiplied by a function of u (unemployment), and z (a ‘catch-all’ variable covering various other factors at play in the labour market).
Lets look at those in more detail:
Expected prices – workers and firms will have a general idea in their minds about the level of expected prices, in other words they will factor inflation into their wage bargaining. If for instance your wages last year were £20000, and you know inflation is 3%, then you know that an offer of £20600 this year will basically be worth the same in real terms. If inflation is 5% then an offer of £21000 this year will be worth the same in real terms. You need to know this before you start wage bargaining so that you can have an idea of your starting point. Of course the firm will know this as well, if the workers try arguing for a 5% pay rise ‘to keep pace with the cost of living’ and inflation is actually only 2%, the firm will tell them to forget it. This concept that expected prices form part of wage bargaining is crucial to the understanding of how inflation persists and how it is hard to get rid of it when it starts to accelerate.
Unemployment – basically as discussed above, high unemployment gives more bargaining power to firms, low unemployment gives more bargaining power to workers. So wages will depend negatively on the unemployment rate.
The catch-all variable, z – this will include things like the levels of unemployment benefit and welfare legislation (higher benefits increase the reservation wage, the minimum wage at which workers are willing to accept work), or the level of trade union power which will also influence workers’ bargaining power. Basically anything other than unemployment, which determines the relative bargaining power of workers and firms, comes into the catch-all.