When to use an open/closed economy model?

General DSGE Modeling Replication Attempts
1

When studying issues like features of optimal monetary policy for a given economy, a choice must be made between using an open economy or a closed-economy model.

How to make this choice? On one hand, you can say that, well, every economy participates in international trade, so no economy is a closed economy. But the degree of openness matters for this choice, right?

My economy is small and commodity-dependent (Ghana). In my recent research, I found that commodity price shocks do not affect the business cycle at all using an atheoretical SVAR model, plus other pieces of evidence. I found that this is plausible because (1) the government sells the economy’s cocoa output in forward contracts to protect the economy from commodity price shocks. (2) Gold companies in Ghana sell in the spot market, however, a lot of the mining firms are foreign-owned, so a lot of the foreign exchange does not come back. And hence there is no big effect of a gold price shock on the exchange rate, inflation rate, interest rate, output, employment, etc.

Moreover, there is less synchronicity between Ghana’s business and that of other economies, regions, and the global economy (even that of its neighbors and trading partners). I found some arguments in the literature that this may be because the agricultural sector is large in African economies, insulating these economies from shocks that emanate from outside the economy. Given the above, should I use a closed economy model or an open economy model when studying features of optimal monetary policy for Ghana? Maybe, there is no definite answer, but, it looks like a closed economy to me in terms of trade (more than an open economy), in the sense that foreign shocks do not affect the business cycle via the trade sector.

Also, Ghana is not an economy where a lot of the citizens trade in financial assets to smooth consumption. The government, however, does borrow from abroad. So when the Ukraine war started, for example, prices went up in major advanced economies, interest rates then went up, and foreign investors abandoned the foreign bonds issued by our government. It resulted in a strong depreciation of the exchange rate, resulting in an inflation crisis.
Can I use a closed-economy model to model Ghana’s business cycle, and capture external shocks (that cause inflation) as inflation shocks or supply shocks hitting the Phillips curve? Or what model assumption would be appropriate in this scenario? I am studying features of optimal policy for Ghana. Thanks for help!!

2

That’s hard to tell in general. It’s a matter of realism vs tractability of the model. From what you describe, the open economy dimension seems to not be of first order importance for the conduct of monetary policy. In that case, a closed economy approach might make sense.

1 Like
4

Dear Prof. Pfeifer, many thanks for the reply. If you don’t mind, may I kindly ask one more question? Can we say a model is open even though it has no explicit trade sector? For example, if we use the basic NK model with no trade sector as in Clarida, R., Gali, J., & Gertler, M. (2000), but we interpret the inflation shocks to the Phillips curve as oil price shocks (like the 1970s and 1980s oil price shocks), can we say the model is a closed to trade but not to foreign inflation shocks, for example? I think we can. Or maybe, no? :slight_smile: Is it ok to call a model an open economy model when it has no explicit trade sector but also inflation shocks can be external shocks?

5

In terms of micro-foundations it’s still a closed economy model, although you may give it the interpretation you outline.

1 Like
6

Dear Prof. Pfeifer,

Sorry to disturb you a little about my personal project:). I got the following response from a reviewer. I think it makes my choice of a closed economy model over an open economy model even stronger. No?

“You conclude from the analysis that the closed economy model is the appropriate model to describe the Ghanaian economy, rather than the open economy model, which has traditionally been considered appropriate. However, you may have concluded too quickly. This is because your results that Ghana’s economy is not driven by trade balance, capital flows, and global and regional demand can be derived from alternative assumptions. That assumption is that the Ghanaian economy is better described by a neoclassical rather than a Keynesian model. In the neoclassical model, the aggregate supply curve (Phillips curve) would be vertical so that demand factors would have little effect on the level of GDP, so it would not be surprising if the results of the analysis you reported were obtained. Moreover, you found that Ghana’s Phillips curve is nearly vertical. So, I wonder why you concluded that Ghana’s economy is closed.”

I think the decision to use a closed or open economy model should be based on whether variables (like trade balance, capital flows, and global and regional demand) affect the economy, and not whether the aggregate supply curve or Phillips curve is vertical or not, yeah? Thanks.

7

It is definitely incorrect that demand shocks don’t play a role in RBC models. For example, government spending shocks also have important consequences for GDP in standard neoclassical models. That is therefore not a justification for the use of a closed economy model.

1 Like
8

Many thanks for the reply, Prof. Pfeifer. If I may ask, that can’t also be a justification for an open economy model, right? The comment seems to be linking the choice between neoclassical and Keynesian model to whether a closed or open economy model should be used. But if I understand, such a link cannot be made, right? Thanks.

Besides checking whether or not foreign shocks matters, is there any other way to justify using a closed or open economy model?

9
  1. Yes, that is correct.
  2. In principle, yes. But often the openness of an economy with respect to goods and assets already provides an indication.
1 Like