When It Doesn’t Matter, and When It Does

One learns macroeconomics by learning a bunch of “it doesn’t matters.”  And then learning when they do and don’t apply.  Here are a few:

1.   It doesn’t matter whether you have a 10% wage tax or a 10% consumption tax.

2.  It doesn’t matter if you increase M by 10% as all nominal quantities rise by 10%, except the nominal interest rate, which is unchanged.

3.  It doesn’t matter if you depreciate your currency by 10%, because your price level will rise to offset any advantages to exporters.

4.  It doesn’t matter who is legally liable, the parties will negotiate the same (efficient) outcome either way.

5.  It doesn’t matter whether you buy a one year bond paying 3.0% or a two year bond paying 3.4%.  Because both have a 3% expected return over the next twelve months.

6.  It doesn’t matter if Korea puts a 10% tax on imports and a 10% subsidy on exports; their effects cancel out.

7.  It doesn’t matter if you can earn a higher interest rate in Australian banks than Japanese banks, as the expected gain is offset by an expected depreciation in the Aussie dollar.

8.  It doesn’t matter if Europeans exempt exports from their VAT, as their exchange rate rises to offset any advantage to exporters.

9.  If doesn’t matter if the trend rate of inflation rises, as the steady state nominal interest rate rises in proportion.

10.  Tax cuts don’t spur demand, because they leave the public no richer than before—hence they are saved.

11.  It doesn’t matter whether the payroll tax is put on workers or employers.

Micro has lots of quasi “it doesn’t matters,” where it doesn’t matter in the way the average person might expect:

1.  It doesn’t matter if you require companies to offer worker benefits; they’ll just pay lower wages to offset the cost.

2.  It doesn’t matter if you force banks to charge lower fees in one area; they’ll just charge higher fees in another.

3.   It doesn’t matter if you require jobs to be safer, or products to be more durable, companies will simply adjust the wage/price.

In these micro cases it does matter for allocative efficieny, but that’s another story.

The various macro neutrality conditions form a beautiful theoretical structure.  Unfortunately, some of them are not always true.  The trick is to figure out when and where they apply.

Most people approach macro from the opposite perspective.  They’ll start with the exceptions, and then gradually those exceptions will take over their entire mental universe.  Here’s an example:

Because wages and prices are sticky in the short run, a sudden increase in the money supply will often depress nominal interest rates for a period of time.  If wages and prices were completely flexible, monetary policy would have no effect on interest rates.  Indeed this has been proven via a special type of monetary policy shock called a “currency reform.”  In a currency reform the government might take 100 old pesos and convert them into one new peso.  The money stock falls by 99%.  Importantly, they also make all wages and prices completely flexible at this point in time, by executive decree.  Thus a worker employed on a 120,000 peso a month labor contract, automatically slides over to a 1200 new peso a month labor contract.  There is no effect on unemployment, and no effect on nominal interest rates.

The majority of people see monetary policy very differently from me.  For them the side effect becomes the policy itself.  Changes in nominal interest rates are no longer a temporary effect of money supply changes in the face of sticky prices, but the policy itself.  This gets them in trouble in the longer term, as the neutrality conditions do eventually hold, and cause the interest rate-focused person to become hopelessly confused.  Countries fall into deep depression and deflation “despite” the ultra easy monetary policy of near-zero interest rates.  Or countries fall to succeed in international trade, “despite” repeated devaluations of their currency.  And so on.

Don’t get me wrong, the non-neutrality conditions are very important in the real world.  Negotiations do require transactions costs, hence liability often matters.  But that doesn’t mean the Coase theorem tells us nothing.  It correctly tells us that there is no rationale for regulating workplace smoking, for instance.  And wage/price stickiness is very important, indeed in my view it explains the Great Contraction (but not the slow recovery.)  But it’s not so important that it produces a permanent trade-off between inflation and unemployment (as we saw in the 1970s.)   And non-traded goods are very important, which explains why PPP doesn’t hold between the US and China.  But they aren’t so important that they prevent rising inflation in China from increasing its real exchange rate, when the Chinese government is too slow to adjust the nominal rate.

Macro is all about balancing the “it doesn’t matters” with the “it does matters.”  Let me give you an example of where I used this framework today.  Nick Rowe has a very interesting post arguing that if we are in a liquidity trap and if the Fed is pegging rates, then higher default risk on T-bonds is expansionary, as it lowers the risk-adjusted cost of private borrowing.  Private debt with the same risk as the now risky government debt sees its yield fall to the rate on T-bonds.

I used the expectations hypothesis to try to refute Nick’s argument.  I’m still not sure I’m right, but here’s my thought process.  The expectations hypothesis of the yield curve says the expected return from holding a 10 year bond for one year, should be the same as the expected return from holding a one year T-bill.  But if the T-bill yield is near zero, then the expected one year return on T-bonds is also near zero, despite their 3% yield to maturity.  So if a 1% default risk it added on, (as Nick assumed) the expected one year yield goes negative.  Now everyone sells all the T-bonds to the Fed, and the Fed must buy because Nick assumed they hold rates steady.  You are left with nothing but cash in circulation, and there is no relative gain for private sector debt, because cash has always been a zero yield/zero default risk asset.

Now I know some of you commenters are chomping at the bit to tell me that the expectations hypothesis of the yield curve is “false” and has been “disproved.”  But that’s where it’s important to always keep in mind what matters, and what doesn’t.  Because even if you are right it doesn’t help Nick’s argument.  That’s because if T-bonds and T-bills are not close substitutes, and if T-bonds yield 3%, then you are not in a liquidity trap in the first place.

I’m not saying that my reasoning is necessarily right here.  But I thought it might be interesting for you to see how my mind works on these problems.  I start with all the beautiful neutrality conditions, and then start considering where they might not hold, for how long they might not hold, how empirically important the non-neutralities are, and what sort of implications they have for the rest of my theoretical edifice.  You need to balance a lot of different factors, and I’m sure I often make mistakes when doing so.  When you read other economists, you can see the ones that tend to have this theoretical edifice in the back of their minds, and those who don’t.  I won’t name any names.

About Scott Sumner 491 Articles

Affiliation: Bentley University

Scott Sumner has taught economics at Bentley University for the past 27 years.

He earned a BA in economics at Wisconsin and a PhD at University of Chicago.

Professor Sumner's current research topics include monetary policy targets and the Great Depression. His areas of interest are macroeconomics, monetary theory and policy, and history of economic thought.

Professor Sumner has published articles in the Journal of Political Economy, the Journal of Money, Credit and Banking, and the Bulletin of Economic Research.

Visit: TheMoneyIllusion

Be the first to comment

Leave a Reply

Your email address will not be published.


*