Real-time estimates of potential GDP: An important, new paper from the Full Employment Project

January 31st, 2018 at 7:43 am

You ask me, the important DC event of the moment wasn’t last night’s State of the Union address. It’s the far less scrutinized meeting going on at the other end of town, over at the Federal Reserve.

Later this afternoon, the Fed’s interest-rate-setting committee will release their monetary policy statement. They are widely expected to pause this month in their “normalization” campaign, i.e., not further raise the interest rate they control.  But their statement will likely reinforce their view that the economy is at full employment, its resources are fully utilized, and their policy thrust will continue to shift from achieving full employment to maintaining price stability. Simply put, more rate hikes are forthcoming.

Their view is not broadly supported by the inflation data, as I’ll show in a moment, but according to an important new paper released this morning by CBPP’s Full Employment Project (FEP), the Fed’s perspective, along with that of other influential economic institutions, like the CBO, has a more fundamental problem: potential GDP is higher than they think it is.

Potential GDP is the level of economic output produced by an economy at full utilization. If you think of the economy as a water glass, full employment implies the glass is filled to the brim. Now, if you’re the Fed, to avoid a spill (inflation), you must stop pouring when you think you’re at the brim. But what if the glass is bigger than you thought? Then the risk is that you’ll stop pouring too soon, at great cost to those who haven’t yet had a drink!

Our new FEP paper, by Olivier Coibion, Yuriy Gorodnichenko, and Mauricio Ulate (CGU) makes precisely that latter case:

“CBO’s and other similar estimates of potential output are too pessimistic, and as such, they encourage policymakers, such as those at the Federal Reserve, to accept lower levels of potential than those which could be achieved. This pessimistic view and associated policies could be extremely costly to U.S. households…Most recently, this has led to some frequently used estimates of potential GDP that are as much as $1.2 trillion, or nearly $10,000 per household, below our preferred estimate.”

What do CGU know that the potential-GDP low-ballers don’t? They borrow a statistical method from a paper by Blanchard and Quah that they claim does a better job separating out temporary and permanent shocks to the economy, thereby getting a more accurate bead on the size of the water glass.

Here’s why that’s so important. Suppose the economy gets whacked by a negative shock like the bursting of a credit bubble. Demand falls sharply and a bunch of people temporarily leave the labor market. A bunch of firms cut back on their investment. Recession ensues. That’s a classic, temporary demand shock. The fundamental supply factors that drive long-term economic growth—labor supply, innovation, the amount of capital per worker—haven’t been permanently thrown of course. Once balance sheets recover and credit flows resume, the economy should make up its losses and revert back to its previous growth rate.

Contrast that with a permanent supply shock, like the one with which we and some other advanced economies are currently dealing: an aging population. Absent a big increase in immigration flows (now we’re back to Trump’s speech), that lastingly slows the long-term growth of the labor force and thus lowers potential GDP.

The problem that CGU document is that current statistical methods employed by the Fed, CBO, and others conflate these two types of shocks, often mistaking temporary downgrades for permanent ones. That, in turn, leads policy makers, like those meeting across town as we speak, to underestimate the size of potential GDP. There’s more room in the glass than they think.

To be clear and fair to all, no one, including CGU, knows the actual size of the glass, which is growing and shrinking all the time in ways that are beyond our capacity to accurately measure, especially in real time. So, while their method is an improvement over the dominant ones in use today, we must be humble about our ability to accurately measure potential.

This is one of the points emphasized by Olivier Blanchard (recall that he was one of the progenitors of the method used by CGU) who was kind enough to provide the FEP with a short comment on CGUs paper. Blanchard writes:

“The basic point of the note by Coibion et al is an extremely important one. Current methods of estimation of potential output do not distinguish between different sources of shocks behind output fluctuations…[CGU’s analysis] is clearly an improvement upon existing methods…But there are limits to it.  Some of these limits can be addressed, by looking at more variables, and so on. Some not so easily. In short, it is a better tool, but it is not a magic one. It must be added to, not replace, the existing panoply.

To me, the best tool remains the inflation signal, at least as far as the labor market, and unemployment, goes. Inflation is the canary in the mine, and a very reliable canary. If the labor market is too tight, if unemployment is below the natural rate, workers will ask for higher wages, firms will be willing to offer higher wages either to keep them or recruit new ones, firms will start increasing prices in order to cover higher marginal cost, etc.”

Which brings us full circle back to the Fed’s meeting. The last reading of their preferred inflation gauge—the core PCE—was up 1.5 percent in 2017, another in a multi-year series of downside misses re their 2 percent target. Some gauges of inflation expectations are tilting up some, and market interest rates have nudged up a bit too, but none of the indicators are signaling serious price pressures. I agree with Blanchard re the limits of CGUs or anyone else’s statistical estimates of potential GDP. But the fact is that the inflation record is strongly consistent with their findings.

We must, therefore, consider the possibility that the glass is bigger than we thought it was, that current methods conflate temporary with permanent shocks, and that there’s more space between actual and potential GDP than the Fed thinks. Closing that gap could make a world of difference to those who are still thirsting for the benefits of the current expansion to come their way.

Print Friendly, PDF & Email

10 comments in reply to "Real-time estimates of potential GDP: An important, new paper from the Full Employment Project"

  1. Robert Palmer says:

    A comment on just one aspect of the post. Why must workers always be the last to get to the punch bowl, usually just as the Fed snatches it away? Over many economic cycles this has depressed wages even as the general level of prices has increased.

    We have a severe housing shortage out here, and a shortage of workers. But local wages are low enough that neither SFR or multi- can be built for a price workers can afford. A medium sized city is mulling writing a check with lots of zeros to a developer of multi-family housing, a gift to him that buys down his cost of construction to a point where he can get his usual return on investment while charging a rent low enough for local workers to pay. If wages were higher that would not be necessary.

    The Great Recession was a long time ago. We need to risk some wage-driven inflation, with wages rising faster than other things, to get incomes back in line with costs. There will be no cost-push knock-on spiral this time. Unions are irrelevant, contracts don’t have COLAs, government is hostile to labor, imports will still be cheap. And the Fed has potent tools to deal with inflation if and when it does get out of hand.


    • Smith says:

      Historical data shows 2 percent inflation target is unsupported, indefensible, an idea from which forms fake news, a lie, a big one. Look at annual inflation going back 50 years. Previous to the Great Recession, when was inflation ever 2 percent or below with unemployment 5 percent or below in a non-recession year? Twice. Once in 1998 because of a big drop in oil prices, once in 1965. Previous cases? Go back to 1956.
      2 percent inflation target is an economic deathtrap that has absolutely no justification, even if the economy were not still trying to recover from the recession and an output gap, even if treated as an average instead of a ceiling.
      Also, who says that inflation should be controlled by raising interest rates to slow the economy and cause greater unemployment. Inflation is not always caused by an overheated economy. Take a moment and absorb what it means. It means other measures should be taken to control inflation. Like what you may ask. Well it depends on the real causes obviously. In our economy, monopoly, oligopoly, foreign cartels, production bottlenecks, bubbles, inequality, over extended credit, government deficits. It’s easy to see the big three auto makers are not constrained by competition since German and Japanese imports cost more. Healthcare and pharma raise prices at will. Colleges charge ever higher tuition. OPEC is restrained only by their overextended pre-fracking age economies. Favorable union settlements and labor gains lead to business raising prices beyond those concessions and wiping out real increases while accelerating inflation. It’s no labor controlling prices.
      Stop saying no need to raise interest rates because wages aren’t increasing.
      Stop that, it’s very wrong and anti labor.
      Stop ignoring the real reasons for inflation.
      Stop pretending like other measures to combat inflation don’t exist.
      Stop and look at what economists and history tells us about controlling inflation.
      Stop acting like the Volcker Carter Reagan Recessions are the only game in town.
      Even in an overheated economy


      • Procopius says:

        “Also, who says that inflation should be controlled by raising interest rates to slow the economy and cause greater unemployment.[?]” The people who make large contributions to politicians do. They invest a lot of their wealth in fixed income assets, i.e. bonds.


  2. Roger C says:

    What is a good argument that the present employment-population ratio for prime age workers should be regarded as permanently confined by a new lower ceiling at the present level? https://fred.stlouisfed.org/graph/?id=EMRATIO,LNS12300060 Still below 2007 and 2.6 points below the peak in 2000. That alone should make us suspicious of a claim that GDP growth has maxed out.


  3. Smith says:

    The population of 25 to 54 year olds is 128 million by Census Bureau estimate for 2016. This equals about 5 million more than the number of full time employees, making equivalent unemployment estimates easy to calculate. Total laborforce is another 33 million when you add in part time workers and 6 million unemployed. 128/160 = .8 Thus add .8 percent for every 1 percent drop in prime age labor force. This makes our current 4.1 percent unemployment rate more like 5 percent, or if you contemplate the era before 2000 when GDP growth slowed by 1/3, two points pushes it up to a near recession 5.8% unemployment.
    In this regard, Donald Trump was nearer the truth during the election by saying the unemployment rate did not reflect the true state of the economy. Clinton could never bring herself to admit this. You know you’re in trouble when Donald Trump is the more truthful candidate.


  4. Vasja Sivec says:

    Gorodnichenko and co-authors have recently produced some great policy-influencing papers. I have only read the abstract and introduction. I intend to read the rest when time permits. It might be that my comment is already addressed by the authors.

    Paper finds that potential GDP responds to temporary shocks, whereas they expected potential GDP to be pre-dominantly driven by permanent shocks. Recent potential output estimate is heavenly Influenced by temporary negative demand shocks, hence potential GDP is likely underestimated. I wholeheartedly agree with their estimate so-far.

    However, I would be more cautious in basing policy recommendation on this finding. There might be a disconnect between “academic” definition of potential output and the definition intended for policy making. Policy makers use potential output or its counterpart output gap to judge budget stability already in the medium-term (say 5-7 years, depending on a country) as opposed to its long-term stability. Thus the estimate that responds to transitory-but-still-long-lived shocks could be more appropriate for budgetary decisions than an estimate that only responds to permanent shocks.


  5. john says:

    As Krugman pointed out years ago, the economy had no problem supplying the 06 demand. Fed presumably thinks supply went poof. John


  6. Tom Cantlon says:

    Speaking of measuring things, WSJ says CFTC chair Giancarlo wants to change how the swaps market is measured. The new method would, “convert short-term interest-rate swaps to their five-year equivalents, downgrading their risk, and would take into account that opposite sides of swaps cancel out each other’s risk.” Since the size and risk (and interdependent house-of-cards nature of having too much of it) was such a factor in the crash, are you going to have any comments on this?


  7. Smith says:

    Historical data shows agreement with a 2 percent inflation target is unsupported, indefensible, an idea that defines a form of fake news, a lie, a big one. Look at annual inflation going back 50 years. Previous to the Great Recession, when was inflation ever 2 percent or below, with unemployment 5 percent or below, in a non-recession year? Twice. Once in 1998 because of a big drop in oil prices, once in 1965. Previous cases? Go back to 1956.

    2 percent inflation target is an economic deathtrap that has absolutely no justification, even if the economy were not still trying to recover from the recession and resulting output gap, even if it was treated as an average instead of a ceiling.

    Also, why is it presumed that inflation should be controlled by raising interest rates to slow the economy and cause greater unemployment? Inflation is not always caused by an overheated economy. In fact where is the evidence that inflation results from an overheated economy? The most famous case of 1970s inflation was one marked by oil price increases and wage price spirals. The very name for the phenomena experienced, “stagflation” tells you it was not overheating, just the opposite. Oil has continued to play an outsized role in upticks in inflation. The Fed itself causes increased inflation through interest rate hikes as Dean Baker has pointed out. Asset bubbles can be a source of inflation, government deficits. Trade deficits sucking up dollars would tend to suppress inflation as money flows out and is not returned. This is not advocate trade deficits which also suck the life out of our economy, but why isn’t the effect not acknowledged? It’s important because if trade becomes more balanced (which we want) then increased inflation magically reappears for those who hadn’t noted the effect.
    Wages are 15 to 20 percent of costs, so that a nominal increase of 5% would mean inflation of 1% without productivity increasing. Fortunately productivity still usually manages a 1% increase each year, except for 2016 when it was zero (the liberal media seems to have buried this significant story of 2016 zero productivity growth). It means other measures should be taken to control inflation. Like what you may ask? Well it depends on the real causes obviously. In our economy, monopoly, oligopoly, foreign cartels, production bottlenecks, bubbles, inequality, over extended credit, government deficits. It’s easy to see the big three auto makers are sensitive to macro demand, but not constrained by competition since German and Japanese imports cost more. Healthcare and pharma raise prices at will. Colleges charge ever higher tuition. OPEC is restrained only by their overextended pre-fracking age economies. Favorable union settlements and labor gains lead to business raising prices beyond those concessions and wiping out real increases while accelerating inflation. It’s not labor controlling prices.
    Why continue to favor raising interest rates as soon as wages are increasing?
    Is this not basically an anti-labor stance that unfairly blames rising wages for inflation?
    Why do economists ignore the real reasons for inflation, which are rising prices beyond any productivity increase less commodity inputs or cost of money or cost of land? None of these factors depend on wages.
    Where is any mention of methods to combat inflation aside from slowing the economy even though inflation isn’t caused by an overheated economy?
    How is it that the extensive history and analysis of the causes and measures to counter inflation are completely ignored?
    Why do economists agree that the Volcker Carter Reagan use of the interest bludgeon is the only game in town for fighting inflation.
    Is there no one who will speak for alternative policies, and this includes measures even if there was an overheated economy?


Leave a Reply

Your email address will not be published.