Outsourcing Is Complicated

Michael Mandel has convinced me that we don’t measure the contribution of outsourcing properly and, if we did, GDP in outsourcing-heavy industries would be revised downward.

The problem is that our tools really struggle dealing with rapidly falling prices for intermediate outputs.

Ok, let me see if I understand this.

Start by assembling two $5 parts and selling the widget for $15 (cost $10). GDP is $15 with $5 allocated to the seller and $10 to intermediate producers.

Now offshore one part for 10 cents and sell two widgets each for $7.50 = $15. Nominal output should be $7.50 x 2 = $15. Real output, however, is $30 because we have to adjust for deflation.

So how is everything allocated? The domestic producer is unchanged at $5 x 2 because his inflation is flat. The foreign producer is the problem. I suppose that producer still gets $5 x 2 real and $0.1 x 2 nominal output allocation. The domestic seller gets $5 x 2.

Nominal GDP = $15 – $0.20 = $14.80.

Real GDP = $30 – $10 = $20

So that’s how outsourcing increases GDP.

What if we’re screwing up the measurement of foreign-sourced intermediate goods?

Mandel points us to an article by Susan Houseman (the full paper is here) that discusses this. The key point in all this is that inflation indexes are a weighted average of domestic and foreign suppliers. Fine.

But when foreign suppliers are cheaper, domestic buyers will switch. The problem is that the BLS doesn’t reweight its index and capture the massive drop in costs.

Mandel makes two points, one of which I don’t agree with and the other I do (based on my little thought experiment above, there):

First he makes the point that the data on electronics products’ share of nominal GDP is understated. I don’t see this, because I think it’s plausible that falling prices are not exactly offset by increasing quantities. This means that measured output will be lower.

Real output allocated to the outsourced function, though, will be massive once you take all that deflation out of the figures. Currently that output is getting allocated to the sellers, which means they’re getting too much.

And I agree with Mike on that.

In any case, I’m all over the implications of his analysis, which basically supports a pro-supply policies by government.

This entry was posted in data, economics. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s