We use cookies on our website to give you the best online experience. Please know that if you continue to browse on our site, you agree to this use. You can always block or disable cookies using your browser settings. To find out more, please review our privacy policy.

COVID-19 RESOURCES AND INFORMATION: See the Atlanta Fed's list of publications, information, and resources; listen to our Pandemic Response webinar series.


Policy Hub: Macroblog provides concise commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues for a broad audience.

Authors for Policy Hub: Macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

Comment Standards:
Comments are moderated and will not appear until the moderator has approved them.

Please submit appropriate comments. Inappropriate comments include content that is abusive, harassing, or threatening; obscene, vulgar, or profane; an attack of a personal nature; or overtly political.

In addition, no off-topic remarks or spam is permitted.

March 5, 2015

Could Reduced Drilling Also Reduce GDP Growth?

Five or six times each month, the Atlanta Fed posts a "nowcast" of real gross domestic product (GDP) growth from the Atlanta Fed's GDPNow model. The most recent model nowcast for first-quarter real GDP growth is provided in table 1 below alongside alternative forecasts from the Philadelphia Fed's quarterly Survey of Professional Forecasters (SPF) and the CNBC/Moody's Analytics Rapid Update survey. The Atlanta Fed's nowcast of 1.2 percent growth is considerably lower than both the SPF forecast (2.7 percent) and the Rapid Update forecast (2.6 percent).

Table 1: Nowcasts of 2015:Q1 real GDP growth

Why the discrepancy? The less frequently updated SPF forecast (now nearly a month old) has the advantage of including forecasts of major subcomponents of GDP. Comparing the subcomponent forecasts from the SPF with those from the GDPNow model reveals that no single factor explains the difference between the two GDP forecasts. The GDPNow model forecasts of the real growth rates of consumer spending, residential investment, and government spending are all somewhat weaker than the SPF forecasts. Together these subcomponents account for just under 1.0 percentage point of the 1.5 percentage point difference between the GDP growth forecasts.

Most of the remaining difference in the GDP forecasts is the result of the different forecasts for real business fixed investment (BFI) growth. The GDPNow model projects a sharp 13.5 percent falloff in nonresidential structures investment that largely offsets the reasonably strong increases in the other two subcomponents of BFI. Much of this decline is due to petroleum and natural gas well exploration; a component which accounts for almost 30 percent of nonresidential structures investment and looks like it will fall sharply this quarter. The remainder of this blog entry "drills" down into this portion of the nonresidential structures forecast (pun intended). (A related recent analysis using the GDPNow model has been done here).

A December macroblog post I coauthored with Atlanta Fed research director Dave Altig presented some statistical evidence that in the past, large declines in oil prices have had a pronounced negative effect on oil and mining investment. Chart 1 below shows that history appears to be repeating itself.

Chart 1: Indicators of drilling activity and oil prices

The Baker Hughes weekly series on active rotary rigs for oil and natural gas wells has plummeted from 1,929 for the week ending November 21 to 1,267 for the week ending February 27. The Baker Hughes data are the monthly source series for drilling oil and gas wells industrial production (IP) and one of the two quarterly source series for the U.S. Bureau of Economic Analysis's (BEA) estimate of drilling investment (for example, petroleum and natural gas exploration and wells). The other source series for drilling investment is footage drilled completions from the American Petroleum Institute, released about a week before the BEA publishes its initial estimate of GDP.

Chart 2: Indicators of oil drilling and natural gas exploration

Chart 2 displays three of these indicators of drilling activity. The data are plotted in logarithms so that one-quarter changes approximate quarterly growth rates. The chart makes clear that the changes in each of the three series are highly correlated, suggesting that the Baker Hughes rig count can be used to forecast the other series. The Baker Hughes data end on February 27, and we can (perhaps conservatively) extrapolate it forward by assuming it remains at its last reading of 1,267 active rigs through the end of the quarter. We can then use a simple regression to forecast the February and March readings of drilling oil and gas wells IP. Another simple regression with the IP drilling series and its first-quarter forecast allows us to project first-quarter real drilling investment. The forecasts, shown as dashed lines in chart 2, imply real drilling investment will decline at an annual rate of 52 percent in the first quarter. This decline is steeper than the current GDPNow model forecast of a 36 percent decline as the latter does not account for the decline in active rotary rigs in February.

A 52 percent decline in real nonresidential investment in drilling would likely subtract about 0.5 percentage point off of first-quarter real GDP growth. However, it's important to keep in mind that a lot of first quarter source data for GDP are not yet available. In particular, almost none of the source data for the volatile net exports and inventory investment GDP subcomponents have been released. So considerable uncertainty still surrounds real GDP growth this quarter.

February 23, 2015

Are Oil Prices "Passing Through"?

In a July 2013 macroblog post, we discussed a couple of questions we had posed to our panel of Southeast businesses to try and gauge how they respond to changes in commodity prices. At the time, we were struck by how differently firms tend to react to commodity price decreases versus increases. When materials costs jumped, respondents said they were likely to pass them on to their customers in the form of price increases. However, when raw materials prices fell, the modal response was to increase profit margins.

Now, what firms say they would do and what the market will allow aren't necessarily the same thing. But since mid-November, oil prices have plummeted by roughly 30 percent. And, as the charts below reveal, our panelists have reported sharply lower unit cost observations and much more favorable margin positions over the past three months...coincidence?

photo of Mike Bryan
By Mike Bryan, vice president and senior economist,
photo of Brent Meyer
Brent Meyer, economist, and
photo of Nicholas Parker
Nicholas Parker, economic policy specialist, all in the Atlanta Fed's research department

December 4, 2014

The Long and Short of Falling Energy Prices

Earlier this week, The Wall Street Journal asked the $1.36 trillion question: Lower Gas Prices: How Big A Boost for the Economy?

We will take that as a stand-in for the more general question of how much the U.S. economy stands to gain from a drop in energy prices more generally. (The "$1.36 trillion" refers to an estimate of energy spending by the U.S. population in 2012.)

It's nice to be contemplating a question that amounts to pondering just how good a good situation can get. But, as the Journal blog item suggests, the rising profile of the United States as an energy producer is making the answer to this question more complicated than usual.

The data shown in chart 1 got our attention:


As a fraction of total investment on nonresidential structures, spending on mining exploration, shafts, and wells has been running near its 50-year high over the course of the current recovery. As a fraction of total business investment in equipment and structures, the current contribution of the mining and oil sector is higher than any time since the early 1980s (and generally much higher than most periods during the last half century).

In a recent paper, economists Soren Andersen, Ryan Kellogg, and Stephen Salant explain why this matters:

We show that crude oil production from existing wells in Texas does not respond to current or expected future oil prices... In contrast, the drilling of new wells exhibits a strong price response...

In short, the investment piece really matters.

We've done our own statistical investigations, asking the following question: What is the estimated impact of energy price shocks in the second half of this year on investment, consumer spending, and gross domestic product (GDP)?

If you are interested, you can find the details of the statistical model here. But here is the bottom line: the estimated impact of energy price shocks is a very sizeable decline in investment in the mining and oil subsector relative to baseline and, more importantly, an extended period of flat to slightly negative growth in overall investment relative to baseline (see chart 2).


In our simulations, the "baseline" is the scenario without the ex-post energy price shocks occurring in the third and fourth quarters of 2014, while the "alternative" scenario incorporates the (estimated) actual energy price shocks that have occurred in the second half of this year. These shocks lead to a cumulative 8 percent drop in consumer energy prices and a 6 percent drop in producer energy prices by the fourth quarter of this year relative to baseline. By the fourth quarter of 2017, 2 percentage points of these respective energy price declines are reversed. In chart 2 above, each colored line represents the percentage point difference between the "alternative" scenario and the "baseline" scenario.

As for consumption and GDP? Like overall investment, there is a short-run drag before the longer-term boom, as chart 3 shows:


So is the recent decline in energy prices good news for the U.S. economy? Right now our answer is yes, probably—but we may have to be patient.

Note: We have updated this post since it was originally released, clarifying a sentence in the paragraph above chart 2 and providing the data for the charts. The original sentence stated: But here is the bottom line: the estimated impact of energy price shocks is a very sizeable decline in investment in the mining and oil subsector and, more importantly, an extended period of flat to slightly negative growth in overall investment (see chart 2).

July 8, 2013

Let’s Talk about Oil

Given its role in touching nearly every aspect of life across the globe and given the higher and volatile prices over the past half-decade, oil supply has been an incessant topic of conversation for much of our recent memory. Yet the tone of the conversation has dramatically pivoted recently from arguments about whether peak oil or sky-high oil prices could spur a global economic meltdown (anyone remember 2008?) to the shifting energy balance as a result of rapidly growing oil production from North America.

Chip Cummins and Russell Gold recently published a piece in the Wall Street Journal discussing how new supply from U.S. shale oil and Canadian oil sands is helping to steady global oil prices.

Crude prices have remained remarkably stable over the past year in the face of a long list of supply disruptions, from Nigerian oil theft to Syrian civil war to an export standoff between Sudan and South Sudan. The reason in large part is a thick new blanket of North American oil cushioning the markets.

This chart helps demonstrate how quickly the oil landscape in the United States has indeed changed. The U.S. Energy Information Administration (EIA) expects national crude oil production to exceed net oil imports later this year, marking a rapid turnaround from the trend of ever-increasing reliance on imports.

However, despite the increase in U.S. oil production, global oil prices have stabilized at relatively high levels, as the chart below shows.

However, the two seemingly opposing narratives—that of high oil prices and that of an emerging oil and gas abundance—are fundamentally linked. In fact, if it hadn’t been for such high oil prices, this new surge in North American oil production may not have happened. It is much more difficult to rationalize drilling activity in deep offshore areas, hard shale, or tar sands—from which, by nature, oil is expensive to produce—without high oil prices. (West Texas Intermediate, or WTI, oil averaged $31 per barrel in 2003, which, even in real terms, is only about 2/5 of today’s prices.) Analysts at Morgan Stanley estimate that the break-even point for Bakken (North Dakota) crude oil is about $70 per barrel and that even a price of $85 per barrel could squeeze out many of the unconventional producers.

What does all this mean for prices? Well, keep in mind that oil is a global commodity. So the roughly two million barrels of oil per day that have entered the market from the U.S. fracking boom represent a big shift domestically but only just over 2 percent of global oil consumption.

And while the United States is seeing growing oil supplies and moderating demand, a different trend is taking place globally, with rising demand from China and other emerging economies coupled with declining supply from older fields and OPEC efforts to keep prices higher through production limits.

However, not everyone believes that higher prices are here to stay. Some analysts have begun to warn that a price crash may be looming. Paul Stevens, an energy specialist with Chatham House, argues that we may be headed for a replay of the price crash in 1986 when high prices triggered demand destruction while bringing new, more expensive sources of supply to the market from the North Sea and Alaska.

Only time will tell where global oil prices will ultimately shake out, but for now, the larger supply cushion has certainly been a welcome development in the United States. Back to the Wall Street Journal article:

The new supply...is acting as a shock absorber in a global supply chain that pumps 88 million barrels of oil to consumers each day. That helps everyone from manufacturers to motorists, by steadying fuel prices and making budgeting easier.

Photo of Laurel GraefeBy Laurel Graefe, Atlanta Fed REIN director, and

Photo of LRebekah DurhamRebekah Durham, economic policy analysis specialist at the New Orleans Branch of the Atlanta Fed

Authors’ note: We didn’t touch on the difference between WTI and Brent oil prices in this post, despite the fact that the changing global oil production landscape has undoubtedly contributed to that spread. For those interested, we recommend some recent analysis from the Energy Information Administration on the narrowing spread between WTI and Brent.