about me
art
biz
Chess
corrections
economics
EconoSchool
Finance
friends
fun
game theory
games
geo
mathstat
misc
NatScience
... more
Profil
Logout
Subscribe Weblog

 
Today I held a presentation (pps, 803 kB) based on Aghion and Howitt's seminal article A Model of Growth Through Creative Destruction (Econometrica, 1992). The presentation was optimized for an online audience, i.e. you'll understand the whole model by just going through the slides. During the presentation I also mentioned Aghion's speech "Growth and Education: Gerschenkron meets Schumpeter". I blogged about it here.

PS: There is one important derivation you find neither in their 1992 article nor in their textbook Endogenous Growth Theory. Since I have written it on the blackboard during the presentation I'll replicate it here:

Expected welfare is
agh01
Given that the innovation process is Poisson with parameter λφ(n) and the stock of technology At grows geometrically we have
agh02
"Factoring out" multiplicative constants yields
agh03
Since
agh04
and ∫e-axdx = 1/a we get
agh05
benny (guest) meinte am 27. Apr, 22:20:
having this in growth theory...
..this semester. we are still stuck with solow and ramsey modell at the moment though. it's like doing this two modells over and over again in every frikkin course. ;) 
HedgeFundGuy meinte am 27. Apr, 22:25:
So what's the intuition of this model's result? In words, what does it say? 
Mahalanobis antwortete am 28. Apr, 17:43:
The
model is a nice finger exercise for more advanced models. Ending up with stationary perfect foresight [sic] equilibria in which log[GDP] follows a random walk with drift isn't too exiting. The innovation process is Poisson and that's it. So all this human capital related stuff like learning, imitation, et cetera isn't incorporated. The only interesting thing is that the laissez-faire growth rate can be larger than the growth rate under a social planner. Though obvious empirically, you wouldn't expect that from a theoretical model.. ;-D. 
bob (guest) meinte am 28. Apr, 11:20:
patent law
Does the model say something about the optimal design of patent law to maximize growth? 
Mahalanobis antwortete am 28. Apr, 17:43:
Nope
stxx meinte am 1. May, 03:22:
What is the arbitrage condition for in the model? 
Mahalanobis antwortete am 1. May, 15:19:
To
determine wages and, therefore, the amount of labor in research/manufacturing.

Both the intermediate good sector and the R&D sector use skilled labor. Saying that the value of an hour in manufacturing equals the expected value of an hour in research seems intuitive.

In my presenatation and in the AH book we have φ(n) = n (i.e. φ'(n) = 1) and the arrival rate is λφ(n). In the Econometrica-article they use φ(n,s), where s is the amount of specialised labor. A firm maximizes expected profits from research by choosing n and s.

λφ(n,s)Vt+1 - wtn - wsts, . . . s.t. labor inequality constraints

So go and call Kuhn or Tucker for help ;-D. 
stxx antwortete am 3. May, 22:40:
Kuhn from the Kuhn Rikon Grill Chef skewer set? 
james (guest) meinte am 9. Nov, 21:53:
The presentation that you offered here was so good. So much that you can get from this. The insight that you offer will give you so much. Great job on this. http://storageandmovers.wordpress.com/ 
santa225 meinte am 10. Oct, 08:40:
Thanks to you besides to the persons who posted this type of educational information. Your ideas are practical to math homework help it is my fundamental date using this welfare so I do hardly distinguish exact course. Hopefully this is the path of the prospective moreover many of persons can benefit. Thank you very.