Does a VAT promote exports?

by on September 30, 2016 at 12:58 am in Economics, Uncategorized | Permalink

Joel Slemrod goes through the details (pdf), including a discussion of a co-authored 1990 paper by Feldstein and Krugman, and also the Lerner theorem.  As you might expect, there is much wisdom, and microeconomics, in that six-page piece.  His answer on the tax side is basically “no,” a VAT does not favor exports, consistent with what Krugman argued a few days ago.  I should note this is not such an easy question, and I don’t think one economist in twenty, if asked on the spot, would come up with exactly the right answer and why (not an excuse for those who get it wrong with prep and Google at their disposal, though I should add this is one of the most esoteric economic errors I have seen a candidate/advisor make).

Here is a summary passage from Slemrod:

First step, understand why a uniform VAT is equivalent to a uniform RST [retail sales tax]; both levy tax on domestic consumption regardless of where goods or services were produced. Second step, calmly reassure oneself that, as is intuitive, an RST does not favor domestic over foreign production and neither encourages nor discourages exports or imports. This implies step three: that a VAT (like an RST) neither encourages nor discourages exports or imports. If step three fails, return to steps one and two until fully convinced.

Quick: say a country taxes imports and subsidizes exports, how quickly can you see that in a first-order model this ought to be neutral?

That all said, it turns out that Trump and Navarro are (partly) right after all, although probably not for the reasons they might have thought.

In fact it is Feldstein and Krugman (p.12) who best explain why, and this has to do with how a VAT boosts savings (and thus trade surpluses), relative to the USA system of taxation:

The best case for arguing that a VAT enhances competitiveness is not what it does but what it doesn’t do: a VAT, unlike an income tax, does not place a tax on saving. Thus, to the extent that a VAT substitutes for an income tax, it will tend to reduce the current propensity to consume. As many economists have pointed out (see, in particular, Frenkel and Rain 1988), to the extent that a value-added tax that substitutes for an income tax reduces current consumption, it will in turn will tend to lead to a trade surplus in the short run. A trade surplus, other things equal, tends to increase the size of the traded goods sector.

Now I am fine with “going all Don Boudreaux” and believing that more Mexican imports are no problem whatsoever.  If Mexico, because of its tax system, saves more and ends up investing more in the United States, great, even if the measured U.S. trade deficit goes up.  I am happy with that situation, whether or not I believe the correct model is one where all trade accounts fall into balance in some super-long run.

But if you banish I from C + I + G, condemn trade surpluses, and believe that single country moves toward a higher trade surplus drain aggregate demand from the global economy, then actually Trump and Navarro have a point. Fortunately, none of that is my view, so we are back to them being wrong.  Perhaps this is a case of “Aggregate Demand Drain for Me but not For Thee.”

1 Jim Birch September 30, 2016 at 1:33 am

Be that as it may, I think you left Trump behind.

2 Ray Lopez September 30, 2016 at 3:01 am

Seems the entire post is strawman: “understand why a uniform VAT is equivalent to a uniform RST [retail sales tax];” – is this not obvious? If not, why not? If it is obvious, then the paper’s conclusions follow logically. I think if one in twenty economists are confused about this, then it shows how muddle-headed is their thinking.

3 PD Shaw September 30, 2016 at 9:46 am

Since I wouldn’t consider Mexico as having a uniform VAT, I don’t see these steps as being all that relevant. For over twenty years, Mexico set its VAT rates lower along the border. Why? The most likely answer is to encourage foreign investment in export businesses. The rates were made uniform (for now) in 2014, but what was the effect?

Otherwise, Navarro’s argument has more to do with the difference btw/ countries which rely primarily on VAT vs. income taxes. It is better to have income in Mexico for production of goods to be consumed in the U.S.

4 Brian Donohue September 30, 2016 at 8:46 am

You build up to the whole “propensity to consume” thing like it’s not completely obvious from the outset.

The US has run trade deficits with other rich people like Europeans and Japanese consistently for decades, largely owing to differential taxation of consumption versus income in the different areas. The US has always been the leading consumerist society in the world. Is this the punch line? What am I missing?

5 Brian Donohue September 30, 2016 at 8:49 am

I mean, if the US migrated to more of a progressive consumption tax system, I would absolutely expect that to reduce our trade and current account deficits. Right?

6 Brian Donohue September 30, 2016 at 8:54 am

In fact, I love the idea of a progressive consumption tax. Let’s hitch this to trade deficit fanatics and get something done.

I don’t think it would do a whole lot for jobs though, but we can keep that on the downlow for now.

7 Brian Donohue September 30, 2016 at 8:57 am

Also, a progressive consumption tax would be a good way to get our pound of flesh outta feckless Boomers and their $20 trillion debt before they shuffle off this mortal coil.

Of course, those crafty old bastards have somehow put up two of their own in 2016…

8 Brian Donohue September 30, 2016 at 9:05 am

Which reminds me: if Trump gets in, the idea of scaling back the estate tax will mysteriously bubble to the surface, despite its overwhelming political insignificance, because it LITERALLY ONLY AFFECTS THE VERY RICHEST 0.2% IN THE COUNTRY.

I have a picture in my head of a rich old Boomer corpse literally holding up his middle fingers in a last act of defiance as they shovel the dirt on.

9 Ray Lopez September 30, 2016 at 9:06 am

You’ve lost the thread and clearly enjoy talking to yourself, BD.

The thread is about whether VAT is equal to RST or not (it is) and whether it would affect the balance of trade (it would not).

Whatever you’re rambling about is not on point or, unlike my rambles, even interesting, sorry.

10 Brian Donohue September 30, 2016 at 9:35 am

Ray, thanks for taking the time to respond to my comment. To the first BIG point (VAT is a sales tax) the only possible answer is: duh.

To the SECOND big point (tax regime affects balance of trade), you are incorrect. It does. Also duh, actually.

Sadly, I see no place for you to inject Bernanke’s 2002 paper into the conversation, which leaves you doubtless nonplussed.


11 Jamie Youngblood September 30, 2016 at 10:23 am

What tax is the VAT tax replacing and how do we know? If it is replacing a corporate income tax to some extent, then why wouldn’t that advantage exporters in the VAT country?

In the example below:
“A more reasonable assumption is that, before tax, the two countries’ manufacturers are equally cost efficient and both require $20,000 per car to cover their cost (and make a profit). Once we correct that mistake, the apparent disadvantage under a VAT goes away and we are back to the RST case. Specifically, the German manufacturer can sell its car for $23,800 in Germany, from which it receives $20,000. Or it can sell it for $20,000 in the U.S., pretax and posttax (because the U.S. has no VAT). The exact same choices are available to the U.S. manufacturer.”

Wouldn’t the US company still net lower profits due to the corporate income taxes it is paying? And wouldn’t that translate into essentially a higher cost of production?

12 PD Shaw September 30, 2016 at 10:46 am

Part of the Navarro argument is about the VAT rebate that plants in Mexico receive when they export goods. I don’t know if that is similar to the German rule, but if it is, the German manufacture could sell for less than $20,000 if it gets a rebate for purchases made that were subject to a VAT.

OTOH, if a VAT and an RST are equivalent, why doesn’t a RST apply to the U.S. sale? If we are operating under the assumption that one is paid entirely by the producer and the other by the consumer, then they are not entirely equivalent in terms of consumer savings rates.

13 louis September 30, 2016 at 12:41 pm

“Wouldn’t the US company still net lower profits due to the corporate income taxes it is paying?”
1) The German company (or US based retailer) would be liable for some US corp income tax on the sale
2) If Germany has an across the board VAT and the US does not, the price level in Germany should be higher and German employees should demand a higher wage all else equal.

14 Troll me September 30, 2016 at 10:37 am

If you stop with the marginal thinking for a moment, and compare an income tax revenue collection to uniform VAT revenue collection system, it should be obvious that the VAT-only system promotes exports, by virtue of not taxing the production of the product which would have otherwise had income taxes built into the price (effectively serving as an export tax).

Not the worst type of “race to the bottom”, but consider what would happen to considerations of tax systems globally is the USA and China both went to “VAT only” in order to abolish income taxes. Other countries would be largely forced to move in that direction.

In advanced economies, revenue collection systems are essentially in place and probably there would be no huge issue. But what of countries with porous borders where wealth centres, no consumption centres, are the easiest locus to raise taxes?

15 Harun September 30, 2016 at 11:19 am

“Now I am fine with “going all Don Boudreaux” and believing that more Mexican imports are no problem whatsoever. If Mexico, because of its tax system, saves more and ends up investing more in the United States, great, even if the measured U.S. trade deficit goes up.”

Its stunning that economists with supposedly internationalist credentials are fine when Mexico or China break the economic rules and hurt their own citizens because some Americans may benefit. (If free trade is a good idea, its a good idea for all, and you shouldn’t be happy when the Chinese hurt their own citizens.) You want to allow Mexicans to come to the USA because you don’t believe in borders, but you’re fine with Mexico impoverishing its own citizens, because…you do believe in borders.

Secondly, this position seems to be quite happy with price signals being massively distorted. You may imagine that keeping the RMB artificially low back in the day only hurt the Chinese, but in reality, it hurts everyone, because a ton of malinvestment is then done by both sides based on a price signal that is distorted. (For example, China central bank sterilizes its trade suprplus by buying billions and billions of MBS…that’s not going to affect anything?)

16 Lord September 30, 2016 at 1:18 pm

Increased savings due to increased inequality, but Saez offers a different model of optimal capital taxation, one much more likely to be real.

17 BC September 30, 2016 at 7:58 pm

Lerner’s Theorem, and an intuitive explanation of it, doesn’t seem to get enough attention in the trade debate. An import tariff may discourage sale of foreign goods domestically but, because it’s equivalent to an export tax, it also discourages sale of domestic goods abroad. This doesn’t depend at all on retaliatory tariffs. Effectively, an import tariff acts as its own retaliation by simultaneously taxing exports. So, for every domestic industry protected by a tariff when competing domestically, another domestic industry is *penalized when competing abroad*. Intuitively, a tariff on German cars in the US discourages those cars from being sold in the US. However, it simultaneously *encourages* those German cars to be sold instead in Germany, where they represent increased competition for US car manufacturers trying to sell in *Germany*.

Mathematically, in the Lerner’s Theorem proof [], if a tariff makes the relative price of German cars in the US (1+t) more expensive than in Germany:

(p2/p1) = (1+t) (p2*/p1*),

then it must also make the relative price of German cars in Germany (1+t) cheaper than in the US:

(p2*/p1*) = (1/(1+t)) (p2/p1).

(This second equation demonstrates the equivalence to an export tax.)

If trade debates highlighted the fact that discouraging foreign companies from selling their goods in the US only creates more competition for US goods being sold abroad, then perhaps fewer people would mistakenly believe that foreign competition costs US jobs.

Comments on this entry are closed.

Previous post:

Next post: