Scott Sumner  

Did Russia avoid high unemployment with NGDP targeting?

A Story of Signaling... We're Number 10! We're Number ...

I've recently read a few articles that suggest the Russian economy is doing better than expected. The Economist attributes this success (or perhaps less failure than expected) to the adroit management of the head of Russia's central bank:

Russia's economy has been held back for years by corruption and rent-seeking, and more recently by Western sanctions and the low price of oil and gas, the country's main exports. Yet the Central Bank of Russia (CBR) is a model of competent, technocratic policymaking. Since Ms Nabiullina became governor in 2013, the CBR has kept Russia's economy, awful though it is, out of worse trouble.

Screen Shot 2016-05-16 at 3.31.26 PM.png
Russia experienced a very severe real shock when the price of oil and other commodities plunged in 2014-15:

Screen Shot 2016-05-16 at 3.10.17 PM.png
The recession might not appear to be all that severe, but note that these are quarterly growth rates; the annualized growth fell by 4 times as much, say from a trend of about 2% to negative 4%.

I was not able to find any data on NGDP growth, and would appreciate any help in that area. However it's clear that a central bank that was targeting NGDP should have allowed inflation to increase sharply during the recession. That is, they should have done the exact opposite of what the Fed did, when it allowed inflation to fall during the severe 2008-09 recession. And it seems like the Russian central bank did exactly that:

Screen Shot 2016-05-16 at 3.15.47 PM.png
A couple points need to be emphasized. The inflation data does not seem to exactly offset the change in RGDP growth. On the other hand, this is not the theoretically appropriate GDP deflator. I could not find quarterly data for the GDP deflator, but I suspect that this data would show the exact opposite problem from the CPI. Rather than overcompensating, the rise in the GDP deflator may well have under-compensated for the fall in RGDP. That's because Russia devalued sharply during the recession---so the price of imported goods (in the CPI but not the deflator) would have soared, and the price of commodity exports (in the deflator but not the CPI) would have plunged.

In any case, elsewhere I've argued that total labor compensation is a better target than NGDP, for a country that is highly dependent on the export of commodities with volatile prices.

We do have one piece of evidence that monetary policy was effective, the unemployment rate:

Screen Shot 2016-05-16 at 3.26.13 PM.png
It's possible that this data is inaccurate, but note that unemployment did rise sharply during the 2009 recession. So Russia's unemployment data is not like the Chinese data, which almost never changes. It's interesting that unemployment only rose from about 5% to 6% during a very severe real shock to the Russian economy, whereas in 2009 it soared to over 9%

So how did Ms Nabiullina achieve this good labor market outcome?

The crisis of 2008-09, when oil prices fell and the world economy stagnated, revealed that the Russian economy was dependent on flighty foreign hedge funds and retail investors. As they pulled money out, the CBR tried to prop up the value of the rouble, losing over $200 billion of foreign-exchange reserves in a matter of months (see chart). Lending shrivelled across the economy. In 2009 GDP shrank by 8%.

. . .

To maintain reserves when the oil price began to fall, Ms Nabiullina accelerated a plan to allow the rouble to float. It fell by 40% against the dollar in 2015 alone. Propping up the rouble would have been popular, since it would have preserved ordinary Russians' purchasing power, but it would have meant burning through the country's reserves again. Instead the CBR channelled dollars to sanction-hit banks and energy companies, to help them repay external debt. Reserves have also been used to finance the budget deficit. As oil prices recover, so the CBR is again accumulating reserves, with a view to hitting the $500 billion mark once again.

So why is a sharp depreciation in the ruble to be preferred over the 2008-09 approach, which kept the ruble stable for a time? Didn't the high inflation of 2015 reduce living standards? Ultimately living standards depend on production, which depends on employment. It does no good to preserve real hourly wages, if there are no jobs:

The rouble's fall has stoked inflation, as imports have become more expensive. As a result, real (ie, adjusted for inflation) wages have fallen by more than 10% since 2014. (They are still triple what they were when Mr Putin took office in 2000.) Interest rates, which in 2014 were jacked up to 17%, have been the only tool the CBR has used to stem the rouble's fall. High rates also help to bring down inflation, currently 7%, towards the CBR's target of 4%. These decisions have "reflected the capacity of the institution to do what is right for the country regardless of the political situation", says Birgit Hansl of the World Bank.

Such steps have been "painful, but necessary", in Ms Nabiullina's words.

This is what the Fed should have done in 2008-09. If the Fed had allowed inflation to rise slightly as RGDP slowed during the housing bust, then NGDP growth would have been more stable. Workers would have seen a reduction in real wages, but total employment would have been higher. With more people working, RGDP would have been higher, and since RGDP equals real income, Americans would have (paradoxically) had higher total real incomes, despite having lower hourly real wages.

Just to be clear, I don't have good enough data to make any firm claims about the Russian economy. I doubt whether NGDP growth was completely stabilized (it probably fell somewhat) and in any case, I do not believe that that stable NGDP growth is the exact optimal solution for a commodity exporter like Russia. NGDP growth should slow somewhat when prices plunge in a big commodity exporting sector. But whatever they did, they ended up with the sort of countercyclical inflation you'd like to see, and the unemployment problem seemed to be much less than you'd expect after such a big real shock.

If Trump wins, maybe he should ask his good buddy Putin for some advice on monetary policy.

PS. Also recall that monetary policy is just one factor influencing an economy, and not the most important. Russia's supply-side environment is still lousy:

Nonetheless, the long-term economic outlook is poor. Ms Nabiullina's critics say the CBR's tight monetary policy is the culprit, since it cripples investment. But corporate profits rose by 50% last year as the rouble value of foreign earnings jumped; companies have plenty of cash to invest. In regular surveys, manufacturers cite policy uncertainty, not high interest rates, as a big constraint. Ms Nabiullina agrees. "Our economic downturn is mostly the result of structural factors," she says. What worries her most is not protracted low oil prices, but "how quickly and dynamically" Russia can improve its business environment. Until then, the CBR will have an outsize role in keeping the Russian economy going.
Update: I found some NGDP data at the IMF, but unfortunately only annual data. In the 2008-10 period NGDP growth fluctuated wildly, from 24.1% in 2008 to minus 6.0% in 2009 to 19.3% in 2010.

This time around things were more stable. NGDP growth slowed from 9.6% in 2014 to 3.2% in 2015, to a predicted 5.8% in 2016. Whereas RGDP plunged by 7.8% in 2009, it only fell by 3.7% in 2015. As I indicated, it is appropriate for there to be some NGDP slowdown in a commodity exporting country, when global commodity prices plunge. Thus the NGDP numbers for 2014-16 don't look all that bad to me. I wonder how total labor compensation did over this period? I suspect that variable was even more stable.

Comments and Sharing

COMMENTS (17 to date)
David R. Henderson writes:

Excellent post.

If I understand you correctly, though, and I may not, there’s a problem with this sentence:
We do have one piece of evidence that monetary policy was effective, the unemployment rate:
This is coming perilously close to begging the question. You admit that you don’t know whether the Russian central bank targeted NGDP. So the drop in the unemployment rate may be due to good monetary policy or it may be due to something else. I tend to agree with you, but it’s not particularly strong evidence for at least your preferred version of monetary policy if you don’t know what their monetary policy was.

John Hall writes:

I agree with David. Great post. I've been watching Russian inflation coming down lately with great interest.

ThaomasH writes:

It's likely that a central bank that was not overly concerned by the short term inflation rate inflation (unlike our dear Fed) or a desire to maintain a fixed exchange rate and also wanted to prevent a fall in real economic activity -- a dual mandate -- would act in a quasi-NGDP targeting way without that being an explicit target and without the need of an NGDP futures market in which to intervene.

Jay writes:

With such a low GDP per capita (just barely above Mexico), am I wrong to be skeptical that their UE actually got below 5% recently? Is this the reported rate or gathered by some international third party?

MikeP writes:

Ultimately living standards depend on production, which depends on employment. It does no good to preserve real hourly wages, if there are no jobs...

This may well be the sentence that most sells people on NGDP targeting.

We are used to viscerally reacting to mercantilists who worry so much about little green pieces of paper rather than actual capital, goods, and services. But it really is counter-intuitive not to focus on the dollars and value of the dollar.

Here, too, NGDP targeting says not to worry so much about the little green pieces of paper: pay attention to actual capital, goods, and services. If a burst of more dollars means that capital and labor are sustainably maximized, then we're better off with that burst than without it.

The problem is separating nominal shocks from real changes: some capital and labor really should find its current uses creatively destroyed as future production and consumption patterns emerge.

Genauer writes:

I think the answer is rather easy:

RGDP and NGDP are not relevant for the CBR setting the exchange rate, current account is.

Russia must maintain a healthy current account surplus, in order to be not bankrupted as some in the West desired.

And that means, given the oil output practically constant, that the decreased export revenue, being mainly oil and oil like stuff has to be scaled to the appropriate amount of Rubels

In the meanwhile the picture has become a little more complicated, namely
a) Russia actually increasing output
b) Putins counter sanctions, against for example European food, are pretty effective, trade with Germany (machinery) has been cut to less than half
c) the initial inflation shock has mostly run its course

And now it is other oil producers feeling the squeeze:

US, Venezuela, Nigeria, Saudi Arabia

the last once, with the US Dollar, expected to be broken this year, even think paying in IOUs

This is about geopolitical power play

Justin writes:

Scott, would level-targeting domestic demand be a good policy here? or do you think targeting average wages is the best way to go?

slotowner writes:

"If the Fed had allowed inflation to rise slightly as RGDP slowed during the housing bust, then NGDP growth would have been more stable."

So your claiming that the FED did not take actions that tried to devalue the dollar & create inflation like say dropping interest rates to zero & institute QE? Did it channel money to say bust hit housing, banks, & car companies?

The Fed tried to do everything it could to cause an increase in NGDP & preserve employment but failed. Russia has so far been able to keep people employed by currently having a internal devaluation of labor costs & wages due to inflation & devaluation. It looks like they are managing it so far but the recession is not anywhere close to the end yet.

Gordon writes:

Scott, over the last couple of years, there have been various news stories about Russian workers complaining that they have not been paid wages in months. Here's an example:

And the BBC article claims that this is happening in hundreds of Russian companies. While NGDP stability would help to keep unemployment from rising, a lack of wage stickiness due to withheld wages would keep it from rising as well.

Scott Sumner writes:

David, I agree, although my update adds a bit of evidence that suggests policy was closer to NGDP targeting this time, than during the previous recession.

Thaomas, I agree. But ultimately it is outcomes that matter. You could argue that during the Great Moderation the Fed got similar results to NGDP targeting from its dual mandate approach.

Jay, I can't say, but GDP can also be low for "productivity reasons". Indeed in less developed countries like India the main problem is low productivity, not low employment. But again, I don't have any knowledge of the accuracy of that data. Certainly the unemployment rate rose less than in 2009.

Mike, Yes, and NGDP targeting does allow for creative destruction. In the two years after January 2006, housing construction fell sharply, but NGDP and RGDP kept growing.

Genauer, It's certainly possible they were targeting something else, and got lucky.

Justin, If we move away from NGDP, I'd rather go toward total labor comp, rather than domestic demand. Average nominal wages are a possibility, I did a paper proposing that once. But I also see some practical problems with implementation.

slotowner, You said:

"So your claiming that the FED did not take actions that tried to devalue the dollar & create inflation like say dropping interest rates to zero"

That's right. Between June and early December 2008, which is the key period when the economy fell off the cliff, the Fed did not have interest rates at zero. Instead they instituted IORin early October, which is a contractionary policy. The dollar soared by 15% in trade weighted terms. The real interest rates on Treasury bonds (TIPS) rose sharply between June and early December.

In 2009, the Fed did some of the things you mention, but by then it was too late to prevent the Great Recession.

You said:

"The Fed tried to do everything it could to cause an increase in NGDP & preserve employment but failed"

That's completely inaccurate. Read Bernanke's memoir, where he says the Fed erred in not cutting interest rates (which were at 2%) in the meeting after Lehman failed. A fiat money central bank has infinite ability to boost NGDP. Check out Zimbabwe. They could create 100% growth in NGDP.

Gordon, That problem occurs in many developing countries, but I don't see it affecting the level of NGDP, just the composition of income.

HL writes:

Ah, I can get these data. Will be right back.

HL writes:

Here you go. The latest data are from Q4 2015.

HL writes:

And the last batch!

Deflator vs. CPI

NGDP vs. compensation in national income accounts

Hope this helps!

S D writes:

As you note inflation was pushed up by 1) the sharp depreciation of the rouble (40%) and 2) the relatively high share of imports in final demand (23%).

But how could the Fed have replicated this in 08/09? What monetary easing (actual and/or signalled) could have engineered a 40% trade-weighted decline in the world's reserve currency? I just don't think it could have been engineered. Remember demand for roubles is demand for Russian imports. It has no other use. The dollar is used for all sorts of things all over the world as well as importing US goods.

Second, imports account for only 1/6 of US GDP. This means that any depreciation has a proportionately lower impact on producer and consumer prices.

The parallel with Russia is just not valid.

Scott Sumner writes:

HL, Thanks for all of that data. It looks like money was still too tight, but not as bad as in 2008-09.

SD, You said:

"But how could the Fed have replicated this in 08/09? What monetary easing (actual and/or signalled) could have engineered a 40% trade-weighted decline in the world's reserve currency?"

I think you misunderstood my point. I'm not suggesting the Fed should have targeted exchange rates, merely that they should have targeted NGDP, and if they had done so, the dollar would have been weaker than it was. Of course the dollar actually appreciated strongly in late 2008.

One of the dudes writes:

Yes, the RUB devaluation was a good move, given the terms of trade shock (oil px drop). But...

There are ex-ante risks with such policy -- there was a non-trivial risk of a run on confidence in RUB and re-dollarization of the Russian economy. In the presence of USD such risks are not existential -- contracts and trade could continue to be conducted in USD as they were in 1990s. So the left tail risk of collapsing commerce was truncated. The US does not have such luxury. No Martian dollars to fall back on. Gold maybe, but without payments systems etc it is a false currency.

Michael B writes:

Quarterly GDP in current prices (is it the same as NGDP? hope so)
Quarterly GDP deflator: (in % to previous year)

Official data by State Statistics Agency.

Comments for this entry have been closed
Return to top