AD-AS Flashcards
What is the AD curve in the Keynesian model?
To endogenise π, note that from an IS-MP model with the Taylor rule a fall in π leads to lower r by the MP curve which raises Y, giving an inverse relationship between Y and π, drawn as a downward sloping curve in (Y, π) space
When could the logic of an individual demand curve extend to the aggregate economy?
If there is a Pigou effect where consumption depends on the supply of real money balances in addition to disposable income
What are the two ways to get an AD curve?
To get a (Y, π) curve the Taylor rule can be used to show a relationship between R and π
To get a (Y, P) curve use the IS-LM model to find that higher P with fixed M reduces the supply of real money balances which raises eqm r for all Y, shifting LM up which reduces Y
What can shift the AD curve?
Anything that affects the IS-MP eqm
Shifts to underlying expenditure components like C̄ and G shift the AD curve by the same amount in the same direction as they shift the IS-MP eqm
Changes to policy rule parameters, particularly r̄
What can change the slope of the AD curve?
Response of R to π: higher mπ leads to larger change in r for small change in π so AD is flatter
Responsiveness of Y to shift in MP: flatter IS => flatter AD, larger mY => steeper AD
What is the primitive supply curve which gives eqm Y*?
A horizontal line in (Y, π) space at some exogenous π
How does the ZLB affect the AD curve?
ZLB implies a lower limit on r which implies an upper limit to Y, denoted Ỹ = 1/(1-c) (C̄ - cT + I(-πe) + G), so there is a kink in the demand curve where it goes vertically downward at Ỹ
If expected inflation changes, the location of the kink changes accordingly (lower expectations => lower maximum Y)
What is the general form of the Keynesian AS curve?
π = πe + v(Y - Ȳ)
This is the same relationship as the Phillips Curve
What are the three justifications for the Keynesian AS curve?
Sticky prices, sticky wages, and imperfect information
How sticky are prices empirically?
Excluding sales, there is significant stability, with adjustment on avg every 8-11 months
How do sticky prices lead to an AS relationship?
If some fixed proportion s of firms have fixed prices so (1 - s) of firms have flexible prices, the aggregate (log) price level P = sps + (1 - s)pf
Firms want to set (log) prices p = P + a(Y - Ȳ) with a > 0 (output gap relates to marginal costs) so pf = p and ps = pe which assuming sticky-price firms expect no output gap gives ps = Pe which gives a (Y, P) relationship P = Pe + a(1-s)/s (Y - Ȳ) which, when subtracting P-1 from both sides and using log rules, gives the (Y, π) AS relationship with v = a(1-s)/s
How do sticky wages lead to an AS relationship?
Assuming workers bargain for a nominal wage while firms care about a real wage, unexpected inflation reduces the real wage so firms use cheap labour to expand production
Workers bargain to W = ωPe where ω is the target real wage consistent with Classical eqm, real wage is this over P so higher P leads to lower wage
A firm with a Cobb-Douglas production function hires labour until the real wage equals MPL which can be solved to get employment as a function of prices
If Ȳ is the output when P = Pe then Y = Ȳ(P/Pe)(1 - α)/α, then taking the log of this and subtracting P-1 gives the AS relation with v = α/(1 - α)
How does imperfect information lead to an AS relationship?
Prices give producers information about nominal value but not real value as the producer doesn’t know whether a change in price was the result of a change in real value or price level
Producers will have formed some expectations of the price their product should sell for pe and the quantity that should be sold at this price ye, so p > pe with some probability could be increase in valuation or fall in value of money, former means production should increase but latter means production shouldn’t change, in aggregate this leads to Y = Ȳ + α(P - Pe) which can be rearranged to get AS
What are some critiques of the different stories for the AS relationship?
It is not obvious that there should be rigidity in wages or prices (workers willing to move off their labour supply curves and firms willing to produce where MR < MC)
Sticky wage story suggest real wage should be counter-cyclical but the opposite is the case
There is now much more information available so imperfect information story seems less plausible
What do extreme values of v do to AS?
From π = πe + v(Y - Ȳ), it can be seen that v = 0 is the exogenous inflation expectations model, corresponding to all firms having sticky prices or perfectly labour demand or zero probability of the price level deviating from the expected value
From Y = Ȳ + 1/v (π - πe) it can be seen that as v approaches infinity we have the classical model, corresponding to no sticky prices or labour irrelevant to production or zero probability of relative price changing (vertical SRAS coinciding with LRAS)
In general, lower v leaves more scope for SR output volatility