Talk:Solow–Swan model

Change in workforce
Why is this written as L[t+1] = L[t]*(1+gL) instead of L[t+1] = L[t]*(1+n)?

n is mentioned above the section as the population growth rate. —Preceding unsigned comment added by 192.38.116.245 (talk) 14:07, 25 February 2010 (UTC)

No need to explain technological progress
I don't know how somebody can say that the model needs to explain why technological progress occurs. That sounds pretty much like wondering why mankind exists...

No. In some cases you get a lot of technological progress, in others little. What accounts for the difference is an important question.radek 22:25, 20 April 2006 (UTC)

Golden Rule Savings Rate
There needs to be some mention of the golden rule level of savings here that sets the savings rate to maximize consumption. But, unfortunately, no such article on that exists yet. Scott Ritchie 22:02, 28 May 2005 (UTC)


 * Hi, I've added a stub under the name "Golden Rule (growth)". I've chosen this name to match similar names on the Golden Rule disambiguation page. Presently, it is a stub; perhaps you are able to extend it?
 * Cheers, Wragge 23:03, 2005 May 28 (UTC)
 * I moved the page to Golden Rule savings rate, as I believe that to be a more complete article title that helps avoid the need for parenthesis (which also might confuse it with the British thing mentioned on the Golden Rule disambig page.) Thanks for getting the ball rolling :) Scott Ritchie 21:01, 29 May 2005 (UTC)

Solution is wrong
Janpieter 00:25, 12 January 2006 (UTC) The solution of the equotations isn't correct. After the differentiating-part the formula is incorrect. Variables are changed. The d is used for multiple purposes, it isn't clear. The same with the variable t, where is it introduced?

Furter, the images of the formulas are incomplete.

I'll try to work it out, but I'm not sure if I'm capable of that.

Well, it looks more messy than wrong - t is standard usage for time, d generally means derivative. The other d stands for depreciation - generally denoted by a delta but hard to do here. It does need to be fixed up a bit. And if possible put that sucker in continues time.radek

Can anyone tell me which d needs to be changed to δ and which should be replaced by delta?

The whole thing is incredibly disorienting. How do you set up a neo-classical model without defining preferences and endowment, and then link UPenn? The analysis looks like a bad undergraduate explanation where the professor ends up waving his hands magically at the end of the proof. I'll try to find some time to put together something more efficient.


 * This is the Solow version of the neo-classical model, not the Ramsey, which I started here Ramsey growth model but haven't had time to finish. This one too needs to be seriously cleaned up. radek 03:17, 19 May 2006 (UTC)

Good Work
I'd like to congradulate the editors of this article for doing a much better job explaining this model than my professor. -Drdisque 16:28, 16 February 2006 (UTC)

Would like to add to this. Was a quite valuable piece to completing my weekly advanced macro PS set in reasonable time. - student at University of Michigan Nov. 1st, 2010  —Preceding unsigned comment added by 141.211.231.33 (talk) 06:08, 2 November 2010 (UTC)

More Critics
It's not true that the fact that the model ignores distribution effects is a mere critic made by extreme marxists!! I think you should add: The use of agregated capital. The use of a representative aggent. And theres more but I don't have the time!!

There should be at least some mention of Anwar Shaikh's work in the criticism section.

Criticisms of the model
What about the economic situation in Austria? Did they introduce special measures for growth? Which were they? Please mention sources. --Maestral 16:23, 28 April 2007 (UTC)

"Austrian" is a school of thought within economics. Models based on the insights of Schumpeter are generally grounded in the Exogenous growth model, and the second half of the statement "better prospect ... than the later Lucas/Romer models" is referring to Endogenous growth models, not Exogenous models. I am therefore removing the "some critics" line, for its inaccuracy and misleading nature. —Preceding unsigned comment added by 63.231.67.167 (talk) 10:04, 26 May 2010 (UTC)

Lucas' Law
There is a page on Lucas' Law and not having heard of it before I had quick look on the web and in other references and have been unable to find anything about Spencer Lucas and his Law and the page is unreferenced and it all seems a bit odd. It has a link to "exogenous growth". Does anyone here know anything about this Lucas? (Msrasnw (talk) 13:54, 30 March 2008 (UTC))

Poverty Trap
Can somebody add a section on the "poverty trap"? It's been explained to me several times, and I still don't fully get it. I think it is: If there are multiple steady states, countries with a low savings rate get stuck at the lowest steady state, and need a period of high savings to get over it and then will converge to the higher steady state. Or something.

Ramsey model
The term 'Neoclassical growth model' often refers to the Ramsey growth model, in which consumption is chosen endogenously, instead of the Solow model in which consumption is an exogenous, constrant fraction of income. The links and redirects between these pages need to be improved. --Rinconsoleao (talk) 13:23, 22 July 2009 (UTC)


 * They are both referred to as "neoclassical growth model". Perhaps a disambiguation at the top of the page is needed?radek (talk) 12:56, 5 September 2009 (UTC)

This article is highly misleading. The neoclassical growth model is not synonymous with the solow model. If one refers to "the" neoclassical growth model, in most cases this means the ramsey-model, sometimes an OLG model but surely not the solow model. This article needs a major rewrite. 84.73.207.121 (talk) 21:53, 18 May 2011 (UTC)


 * True. For instance, Acemoglu (Introduction to Modern Economic Growth, Ch. 8), refers to the Ramsey-Cass-Koopmans model as "the neoclassical growth model". This article needs to be rewritten. Let's start be moving it to the appropriate lemma. --bender235 (talk) 21:30, 25 November 2013 (UTC)

Technology things
The production function with different technology presentation, we could solve all the growth model with the following different technologies:
 * Labor augmenting (Harrod neutral) technology $$Y=F\left(K,AL\right)=K^{\alpha}\left(AL\right)^{1-\alpha}$$,
 * Capital augmenting technology $$Y=F\left(AK,L\right)=\left(AK\right)^{\alpha}L^{1-\alpha}$$
 * Hicks neutral technology $$Y=AF\left(K,L\right)=A\left(K\right)^{\alpha}L^{1-\alpha}$$
 * embodied technology $$Y=K^{\alpha}L^{1-\alpha}$$, after remove technology from K, $$Y=\left(A\bar{K}\right)^{\alpha}L^{1-\alpha}$$ where $$\bar{K}=\frac{K}{A}$$, then when viewed with Y and $$\bar{K}$$, it is capital augmenting technology.

By the way, which technology was used in Solow's original paper? Is there anyone who has checked this? Jackzhp (talk) 18:39, 4 July 2010 (UTC)

A possible problem with the graphs
I think the first graph should have a k rather than the K as the variable on the horizontal axis. k=K/L so I think as it stands it is potentially misleading. The other graphs also seem to be labelling axis with Y and K even when variables are y and k. (Msrasnw (talk) 14:42, 27 October 2011 (UTC))


 * Yes. Basically you have to copy the graph to your favorite image program and manually change it, then re-upload it to Commons as a derivative work (easiest if you use their "old upload wizard").  Volunteer Marek   10:13, 29 November 2011 (UTC)


 * Sorry I have made a minor pigs ear of the graph with a stry k in the title - still it is better than it was.... I will tidy later (Msrasnw (talk) 18:01, 1 December 2011 (UTC))

I think I have fixed them now (Msrasnw (talk) 00:18, 2 December 2011 (UTC))

wrong
"A key prediction of neoclassical growth models is that the income levels of poor countries will tend to catch up with or converge towards the income levels of rich countries as long as they have similar characteristics – for instance saving rates. Since the 1950s, the opposite empirical result has been observed on average. If the average growth rate of countries since, say, 1960 is plotted against initial GDP per capita (i.e. GDP per capita in 1960), one observes a positive relationship."

The key phrase in the above is "as long as they have similar characteristics". Hence this para is talking about conditional convergence. Then the article goes on to say:

"These observations have led to the adoption of the conditional convergence concept....Evidence for conditional convergence comes from multivariate, cross-country regressions."

So the article contradicts itself. What it is obviously trying to do - very very clumsily is to say that "absolute convergence" doesn't hold but "conditional convergence" does (sort of). But "absolute convergence" is NOT an implication of the model, and especially not a "key prediction" of it.

 Volunteer Marek  10:16, 29 November 2011 (UTC)

You are correct. The presentation was incorrect in several places in its explanation of convergence to the steady state. I have edited the text, and I think I have eliminated the errors. Macroprofe (talk) 16:34, 14 February 2014 (UTC)

There are more than one thing wrong with this model. The most obvious one is the absence of land values from what is a variation on the Cobb Douglas equation for productivity (which has the same fault). Without the use of land no production is possible and the more valuable the land the greater amount of productivity is possible. Adam Smith seems to have been completely forgotten, since land was the first of his 3 factors of production and is not capital. Durable capital goods being the second factor does not vary in value with population density as land does. Also land cannot be made in the way that all kinds of durable capital goods does nor can it depreciate in the same way.

Another erroneous part of the Solow-Swan model is that knowledge is a part of the third factor labor. Skilled workers with suitable knowledge are more productive but their knowledge cannot simply be purchased and immediately applied without it becoming a part of the work-force. It should not be separated from the labor part although it is seen that in the equation it is kept very close to the labor part itself.Macrocompassion (talk) 06:23, 5 June 2017 (UTC)

"Neoclassical growth model"
Neoclassical growth model redirects here, but surely it should redirect to the Ramsey-Cass-Koopmans page, which is only rarely identified as Ramsey-Cass-Koopmans and is almost always called the neoclassical growth model. 128.135.100.102 (talk) 22:10, 12 April 2015 (UTC)