Institute For Regional Forecasting
For Better and for Worse: Oil Industry Still Drives Houston’s Economic Outlook
Houston's economic outlook is again firmly in the hands of the oil industry, and this brings good news and bad. The good news is that the east side of Houston is gripped by a major construction boom for petrochemicals and other energy infrastructure. Built as it is on a premise of cheap natural gas, this boom seems sure to continue and increase in size over the next couple of years. The bad news is that the price of oil has recently fallen below $60 per barrel, and these low prices promise significant cuts in domestic oil exploration and drilling. (Figure 1) These cuts will be felt from Houston’s machine shops and factories to its office towers. The question becomes how this strange mix of good news and bad balances out to affect Houston’s economic prospects in 2015 and beyond.
In the last 25 years, we have not experienced greater uncertainty about where the Houston economy is headed. What are Saudi Arabia’s intentions regarding the price of oil? Do they really have a goal of pushing Texas and North Dakota into the role of swing producer in global oil markets? Now that oil prices have fallen sharply, how will the industry react? We have not been through a downturn like this one, with decisions guided by the new economics of horizontal drilling and fracturing. Can the construction boom in Houston’s petrochemical complex provide shelter from the drilling cutbacks? How much shelter can these temporary, blue-collar jobs really provide?
In the face of great uncertainty about what path Houston’s economy might follow, this article engages in some scenario writing – a couple of simple stories about what might happen in coming months, and illustrating the consequences of these events on local job growth. These stories assume moderate to heavy cuts in exploration and production expenditures, and that the chemical construction boom proceeds at different rates. Both the scenarios show continued job growth in Houston, but only with substantial help from both chemical construction and a healthy US economy. We also raise some important questions about how the results of these scenarios split Houston between a booming blue-collar east side of the metro area, and a white-collar west side that will probably bear much of the pain.
Oil price has fallen steadily over the second half of 2014. Although the Saudis have not articulated their rationale for promoting a low oil price, or at least for passively accepting a low price, their strategy has emerged as one of helping their friends and punishing adversaries. The friends they help include the US and Europe. The weak European is particularly important to the Saudis, and low oil prices work to stimulate what has become their most important market. The adversaries that suffer include ISIS, Iran, and Russia, as well as OPEC’s usual collection of cheaters such as Venezuela and Nigeria.1
If the US is a friend and beneficiary of low oil and gasoline prices, Texas and North Dakota probably make the Saudi list of enemies. These states are home to the three great American shale oil plays — the Bakken, Eagle Ford, and Permian Basin. Figure 2 shows the surge in US domestic oil production since 2010, largely led by the shale boom that has reversed 40 years of declining American oil production. It has moved the US into third place as a crude oil producer, just behind Saudi Arabia and Russia, and if condensate and other liquids are counted, the US is number one. We have reduced the import of crude and petroleum products to about 33 percent of domestic consumption, reaching the lowest share since 1985.2
Since the 1970s, the Saudis have assumed the role of swing producer in global oil markets, stabilizing oil prices by increasing production when prices move too high, and cutting their oil production if crude price fall too far. The surge of relatively high-cost crude produced by the US apparently has forced a rethinking of this strategy: Is it better to cut production and protect price, or to allow prices to fall and protect a dwindling market share? It might cost $65 per barrel to produce oil from the deep waters of the Gulf of Mexico or from US shale, while it costs $6-$8 in Saudi Arabia. Why not make the high-cost US producer bear the burden of cutting production? It is a messier process than the centralized decision-making in Saudi Arabia, but potentially just as effective. Low prices bring less cash flow to US producers, and force capital spending cuts; high-cost producers are driven out of the business; fewer wells are drilled, and production ultimately falls. There are potentially many layoffs and a few corporate failures along the way, but production falls and higher prices are restored over a period of months. And it makes US shale production a cyclical industry – down with every oil surplus, up only as oil prices rise.
This outcome of this confrontation depends on how much exploration and production will be cancelled as crude price moves from $85, to $75, and $65 per barrel. Can the US shale producer outlast the deep pockets of Saudi Arabia’s Aramco and a $700 billion sovereign fund? The Saudi government — depending on $90 per barrel to balance the public budget — will find this a painful experience as well.
Cheap Oil Meets Cheap Natural Gas
Oil price is critical right now in determining the capital spending for exploration and production (E&P) by the American shale industry. The price of domestic natural gas collapsed in late 2011 in the face of high volumes of new gas production from shale, with the fall helped along by a weakly growing US market. About 600 rigs were pulled away from gas-directed drilling, and the rig count fell by 15 percent. By 2012, eighty percent of US drilling was directed to oil, and the bulk of the US shale boom was left resting on the price of crude.
Figure 3 shows that current inflation-adjusted expenditures for E&P have tripled since the 1982 peak, the previous standard for a massive expansion of the oil industry. Between 2003 and 2008, the real domestic E&P spending rose at compound annual rates of 35 percent per year. This rapid growth was broken by the financial crisis in 2009, then again by the 2011 fall in natural gas prices, but E&P spending remains at very high levels. In 2014, producers seemed to put behind them any negative thoughts left from the financial crisis or low natural gas prices, and pushed strongly back into the oil fields. By late 2014 the fracking fleet of US oil service companies was sold out, and expected to remain fully engaged through 2015. The shale oil boom was set to continue strongly – until the price of oil began to slip quickly last fall.
How will the industry react to lower oil prices? We did experience a major downturn in E&P expenditure and oil industry employment in 2008-2009, but that event also included a major financial crisis and the worst recession of the last 75 years. A recent survey of about 100 oil producers by investment house Tudor, Pickering, and Holt might be characterized as telling us that E&P growth would certainly slow as price fell below $80 per barrel, and would likely be reversed below $75. Only 4 percent of respondents indicated that the price they used for forward drilling plans was less than $75, and 75 percent said they would be trimming their drilling plans at $75. As price dips below $60 dollars per barrel – levels not even contemplated by this recent survey — substantial cuts in capital spending seem to be in the offing.
Meanwhile, in east Houston, a major construction boom is underway based on cheap natural gas. Although a wide variety of projects are underway, with some responding to new infrastructure and energy processing needs from the shale boom, the focal point of the expansion is petrochemicals. In contrast to a refiner – who produces oil products such as gasoline or jet fuel – the petrochemical producer in North American uses natural gas liquids to make intermediate products that become plastics and synthetic rubber. For example, the natural gas liquid ethane is processed into the petrochemical ethylene, which then is used to make plastic such as polyethylene or polyvinyl chloride. Outside North America, these plastics are made with oil-based feedstock priced in recent years at $100 per barrel: meanwhile, our natural gas liquids were priced like natural gas at $4.00 per thousand cubic feet, or the oil-price equivalent of $22 per barrel. The result was enormous competitive advantages for the North American ethylene producer from low natural gas prices, and it has set off a race by many companies to enter or expand their presence in a highly lucrative market.
Table 1 summarizes eight ethylene plants that are proposed or under construction for the Gulf Coast region between Corpus Christi and Lake Charles. The large projects on this list, together with the downstream plastics plants associated with them, will each have a price tag near $5 billion dollars. Thousands of workers will be on site as each construction project reaches its peak. A recent assessment by Goldman Sachs (published with the price of oil near $60) put the prospects of completing the Baytown, Freeport, Ingleside, and Point Comfort plants at 100%, and Sasol’s Lake Charles plant at 75 percent. The reports points out that for many other projects that have been proposed, but are which not far advanced, we are likely to see slowdowns and cancellations.3
LNG plants are another major source of regional construction activity, built to liquefy and export natural gas. North American natural gas today cannot be exported to Europe or Asia, and the large quantity of shale gas trapped here is a major contributor to weak natural gas prices. Two plants have completed the multi-agency permitting process (at Sabine Pass and Freeport), and several more are expected to be built in North America. With a construction cost of $10 billion each, these projects are again major sources of work for the Gulf Coast, including the Houston region.
This construction boom is built on low natural gas prices, and the problems that low oil prices pose upstream should not be a broad threat to this construction. In some cases, where high production levels of liquids justify the project – such as export facilities for propane, butane or ethane – there might be a need to rethink future supplies in a time of production cuts. But the ethylene and LNG expansions seem likely to continue. In the past, US ethylene has been built for domestic production, and was only exported opportunistically. The current plants are built as export platforms based on significant cost advantages over foreign producers. Margins may be squeezed now, but looking forward, no one expects the price of oil to remain below $60 for a prolonged period. Ethylene feedstock priced at a conservative $7 per thousand cubic feet, and oil at $80 per barrel, yields a competitive advantage of natural gas over oil that continues at roughly $40 per barrel.
Counting up Local Jobs
Until recently, the energy industry found itself in the unusual situation of facing high oil prices and low natural gas prices. High oil prices set off a shale-based exploration boom; low gas prices inspired a huge surge in building or expanding infrastructure and processing capacity downstream. Until this fall, it looked like the story of the Houston economy for 2015 would shape up as a major collision between competing resource demands within the oil industry – a race between upstream and downstream to tie up blue-collar construction skills and machine shop capacity. Now the more likely question for Houston will be: How well can the downstream construction boom substitute for job losses in the E&P sector?
How should we count up new jobs in Houston? New jobs are created locally by broad trends in US growth or by the expansion of the local oil cluster. Over the very long-run, probably about half the jobs in Houston are created by the national economy and half by the oil industry. Houston’s oil-related employment currently includes about 300,000 employees of oil producers, oil services, machinery, fabricated metal, refining, chemicals, and pipelines.
Economists often describe employment, income, or production as being basic or nonbasic, depending on the industry involved. Basic employment, for example, is found in industries that reach outside the local metro area, and sell into larger regional or national markets. The oil industries listed above are perfect examples of basic jobs, using local talent to serve national and global markets from Houston. But other non-petroleum industries in Houston also tap into broader markets – wholesale trade, air transportation, and some business and professional services, for example.
Figure 4 shows the recent level of both the oil-related base jobs and the non-energy part of the base, with about 55 percent of jobs being oil-related. Over the long-run, we might expect the two to grow at similar rates, but in recent years, the size of the shale boom has combined with weakness in the national economy to create a considerable disparity. Much of the recent surge in jobs in Houston – indeed, what has set Houston’s economic performance apart from the US economy since 2010 – has been the energy boom set off by $100 crude.
Nonbasic employment is inherently local activity that does not reach outside the metro area, and does not drive regional growth. These industries and their workers provide critical services to the community, but are dependent on base industries to create regional income and jobs. Bars, restaurants, grocery stores, dry cleaners, hospitals, real estate development, and drug stores fit this description. Figure 5 shows the number of nonbasic jobs supported by each base job in Houston, typically between three and four. Every time a new job is created (or lost) in energy or in another basic sector, another three or four jobs are created (or destroyed) in the months that follow.
Before we get into the details of various scenarios, let’s set some expectations about the rate of job growth in Houston. First, we estimate that each one percent increase in US employment will create around 33,000 jobs in Houston.4 The impact is felt over time, and the jobs arrive over a period of 8 or more quarters. In all of our scenarios, we are going to assume that the US economy continues to perform well, generating new jobs at 1.7 percent per year. This means that the annual count of new jobs in Houston always begins with a base of about 56,100 (or 1.7 times 33,000).
For the cluster of 300,000 oil jobs in Houston, a one percent expansion can create 18,000 new basic and nonbasic jobs over the following several quarters. Some quick arithmetic can illustrate how quickly swings in the oil market might affect local job growth. If we begin with 56,100 jobs from the national economy, a 3 percent increase in the oil base provides 110,000 new local basic and nonbasic jobs; a three percent decline in the oil base means 2,100 new jobs; a six percent decline in oil employment means a loss of 51,900 jobs. Clearly, things can change quickly in a commodity-based economy like Houston, and falling oil prices have important implications for local job growth.
Cuts in E&P Spending Upstream
There is great uncertainty about even the near-term level of oil prices and the response of oil producers to low prices. For our first scenario, let’s start with a 20 percent cut in E&P capital spending that works on the following schedule: a 10 percent cut in 2015Q1, followed by another 10 percent in Q2; this 20 percent reduction remains in place through Q4. In 2016Q1, E&P spending rises by 10 percent, and then returns to 2014Q4 levels in Q2. By 2016Q3, capital spending begins to grow rapidly again at 8 percent annual rates. Alternatively, consider Scenario 2, which proceeds on exactly the same schedule, but entails a 33 percent reduction in capital spending and 5 percent growth in E&P beginning in 2016Q3.
In both cases, we assume that the US economy grows at 1.7 percent. If there was no help from downstream construction (considered at length below), these cuts are sufficient to induce a mild recession in Houston. Energy-related jobs in Houston fall by 18,000 in Scenario 1, a decline of 6 percent spread over two years. In scenario 2, they fall by 23,600 or 7.9 percent. Total payroll employment in Scenario 1 would see combined losses of 50,000 jobs in 2015-16, while Scenario 2 would mean a 75,000 job decline over the same two-year period. Growth bounces back strongly in both scenarios as E&P spending picks up again in 2016.
A few very important caveats should be mentioned.
- The length and the depth of the E&P cuts are simply examples. What actually happens will depend on the recovery of the global economy, as well as the strength of will exerted by the Saudis and the depth of their pocketbook. And it will depend on the uncertain response of producers in Texas and North Dakota in cutting production.
- Statistics seek to replicate the past and extrapolate past behavior to current circumstances. We have never been through a scenario quite like this in the era of high-cost drilling. Is the past response really a good guide to the future?5
- Will the expected white-collar job cuts materialize? The huge downturn in oil markets in the 1980s left the US with a surplus of oil-related skills, and another generation of oil workers was never trained. Now the baby-boom generation is retiring, and there is a shortage of core skills such as geologists, geophysicists, and petroleum engineers. An effort to replace the retirees has been underway for several years, and companies may hoard these skills in difficult times rather than consider layoffs. Outside the core skills, however, in any support or overhead position, whether inside the energy company or as a contractor, white-collar employment will certainly be a target for reduction.6
- How fast will the American shale industry grow after an oil-shock like this? Does it simply resume the rapid growth of the past? Or does it emerge as the swing producer in global markets, looking at a highly cyclical future – ramping up production as prices climb, but cutting back and winding down production every time surplus oil emerges on global markets?
Downstream Construction Boom
So far, Scenarios 1 and 2 considered substantial cuts in E&P, but not the downstream spending that is expected to continue and grow to high levels on the east side of Houston. Downstream spending is often neglected in terms of its economic impact of the oil industry, mainly because it is usually small compared to the upstream. Spending on refining and chemicals is mostly focused on the Texas and Louisiana Gulf Coast, but often the projects have more meaning to the Baytown or Deer Park economy than to the broad Houston metropolitan area. That is certainly not the case right now, as the expansion of shale-related infrastructure, chemicals and liquefaction plants has created a boom of unprecedented size and scale.
The east-side construction jobs should be considered as part of Houston’s energy base. We want to create a couple of downstream scenarios that include estimates of the number of construction workers associated with these projects. We will then ask how well this construction might offset the loss of basic employment due to E&P cuts.
- We began with an assumption of very large industrywide capital expenditures for shale-related projects. About $90 billion will be spent for shale-related petrochemical plants, or for other basic processing or transportation-related infrastructure. Another $50 billion would be spent for LNG plants in the US. Petrochemical work began in 2014 and continues through 2018, peaking in 2015-16. LNG construction peaks in 2016-17.
- Statistical analysis found a consistent relationship between Houston construction employment and the level of domestic downstream spending. A one percent increase in the industry’s downstream capital expenditure brings 145 new jobs in Houston, based on the current 200,000 local construction workers.7 This seems like a small number, until you realize that we are considering quarter after quarter of double-digit increases in capital spending. The number of workers can and does add up very quickly.
- Not all announced projects will come to fruition. For Scenario 1, we assumed an attrition rate of one-third percent for these projects, and reduced the number of local workers accordingly. For scenario 2, we assumed that only one-half of the projects are built. Current events may play some role in the cancellation of projects. Margins for some chemical plants may be squeezed by low oil prices, where oil is the competitor feedstock. They will be delayed or cancelled. Or some projects that are based on large quantities of future surplus liquids, like export facilities for ethane, propane and butane may need to be rethought. Generally, however, most projects should remain economic, built as they for a 30-year life, and built because of assumed low natural gas prices relative to crude long after the current oil-price decline is over.
- Figure 7 shows the assumed level of capital expenditure for the two scenarios, and the projected number of local workers. The number of workers is large, but not out of line with the numbers of construction workers often cited as required for this push to build plants and other energy infrastructure. The figure also makes clear the temporary nature of construction work, building up to high levels, and then winding down just as quickly. Just as the construction workers are needed to fill in for the loss of upstream activity in 2015-2016, resumed growth in E&P is needed to fill in for the loss of construction work beginning in 2017-19.
We can think of these construction workers as substituting for job losses upstream. Building these new pants is basic activity – drawing money for construction budgets from global corporations and bringing it into the community – and the construction workers are drawn here from all over the US. We should add them to Houston’s energy base.
But do these temporary, blue collar jobs make good substitutes for the jobs lost upstream, many of them highly-compensated white-collar office workers. Looking beyond obvious differences in compensation, we saw in Figure 5 that through employment multipliers a new job in Houston’s economic base will typically support 3 to 4 more jobs.8 It is unlikely that temporary jobs that last for only 18-24 months could support much new home construction, apartments, or most retail and office space. However, the multipliers should work well enough for linkages back into the industrial, wholesale, and transportation sectors that are needed to support construction activity.
To compensate for the temporary nature of these jobs, we replaced upstream jobs in the energy base with construction workers on a 2-for1 basis – two construction workers compensate for the lost salaries and the multiplier effect of a single worker upstream. This was applied to both the expansion and contraction of construction work. Adding the energy employment losses upstream and the energy gains downstream, we come up with a new total for the number of energy jobs in Houston.
Scenario 1 assumes that E&P falls by 20 percent, and that 2/3 of construction work continues; Scenario 2 is a 33 percent cut in E&P, and only 1/2 the downstream construction work materializes. Figure 8 shows how the energy employment rises in Scenario 1 in early 2015, but then turns flat and declines with relatively small losses in 2016 and 2017. In Scenario 2 the number of energy jobs declines further and recovers more slowly, with more conservative assumptions both upstream and down. The recovery is timed differently in the scenarios because Scenario 1 assumes a resumption of E&P growth at 8 percent after 2016, while Scenario 2 assumes 5 percent. However, both scenarios strengthen in 2017-2018 as the return of strong upstream activity outweighs the winding down of construction work.
Applying these energy employment figures to a forecast of Houston jobs, we find that local job growth continues in both scenarios. (Figure 9) The number of new jobs in Scenario 1 (20 percent drop in E&P and 2/3 of construction completed) is show in Table 2 from 2014 to 2018, along with the annual percentage increases. The numbers seem disappointing in contrast to the 100,000 new jobs that Houston has averaged each year since 2011. However, given the likely damage to the upstream energy base from current low oil prices, these figures look very good indeed. Scenario 1 sees job growth that exceeds the 1.7 percent assumed for an increasingly healthy US economy. Scenario 2 (a 33 percent drop in E&P and only half the construction completed) sees job growth fall below the US growth rate but staying positive and above 1.3 percent. The combination of chemical construction in east Houston and a strong US economy keep Houston well out of recession. Look back at Figure 6 to see what happens to Houston’s economy without the chemical and LNG expansion.
As this is written, with WTI crude prices below $60, the best guess is that we will see a substantial hit to upstream E&P in 2015, and that the downstream construction remains relatively healthy. Oil producer strategy at this point must be to cut capital expenditures, reduce wage, salary and other costs to match cash flow, and then sit it out until crude price and cash flow improve.
Using our scenarios, what happens if we combine a 33 percent cut in E&P with an assumption that 2/3 of petrochemical and other construction is finished? Figure 10 shows that this combination pulls the path of energy employment to the low side of the range of scenario outcomes. Conservative assumptions about the impact of construction work play a role here, replacing lost jobs upstream only on a 2-for-1 basis. Also, the 33 percent cut in E&P comes with an assumption that E&P expenditures recover at a 5 percent rate after 2016; this is in contrast to the 20 percent cut and an assumed 8 percent growth in E&P after 2016.
Figure 11 shows that the tendency of the energy base to follow the low end of the range of scenarios carries through to total payroll employment. The third column of the table on the right of Figure 11 shows the job growth associated with the scenario mix of a 1/3 cut in E&P and 2/3 of petrochemical work completed.
Given Houston’s dependence on oil exploration and production, there is never a good time to see oil prices fall as far and fast as they have in recent months. But if we had to choose such a time, our two scenarios indicate that this might be the moment we would have selected. The construction boom on the east side, combined with the growing health of the US economy, point to continued job growth in Houston despite the setback to E&P.
Simply working through the scenarios provides insight into the extent of the uncertainty facing oil markets, the oil industry, and the Houston economy. These scenarios should serve to illustrate how events might play out in these specific circumstances, but should not be taken as a guide to exactly how events will play out over the coming months.
We are still very early in a story that will take many months to come to an end.
Our scenarios focus on a single round of cuts in the E&P industry, and the recovery that follows. Limited damage to job growth was done in this particular case. A more disturbing aspect of this cycle is whether it might be the first of a long series of cycles to come. If US shale is moved into the role of swing producer, similar cuts to US oil production will be needed to rebalance global oil markets with every emerging surplus. The shale industry will ramp up again only when oil prices justify the expansion. And there will probably not be a local construction boom to offset the pain of these future cycles in Houston. The shale industry’s attractiveness to lenders and investors can only be damaged by this cyclical behavior, pointing to more moderate growth in the exploitation of shale over the long term than it has enjoyed in recent years.
We may look back one day at the overall statistics for job growth in Houston in 2015 and 2016, and simply see a period of moderate growth after years of out-sized performance. But the average may disguise substantial differences between the booming blue-collar east side, and the white-collar west side that will bear much of the pain. For example, a steady flow of young college graduates and professionals has come to Houston seeking an urban life style that fueled an upscale apartment boom inside the 610 Loop. This influx of young workers seems likely to be interrupted. Growth in the need for office space and high-end retail also seem likely to slow based on the income, class, and demographic dimensions of job growth in the scenarios we outlined. Industrial activity, however, whether on the ship channel or the machine shops of north and northwest Houston, will continue strongly, as cancelled projects upstream simply open a place in line for on-going chemical construction.
1 Current weak crude prices are due to a variety of causes, with US shale production only one cause. Over production by OPEC members, and heavy sales from Libya and Iran to raise revenue to fight insurgencies, have added to growing US supplies. On the demand side, there is substantial weakness brought on by recessions in Japan and Europe, along with a broad slowdown in emerging markets such as China, India and Brazil.
2 Energy Information Administration, How Much Petroleum Does the US Import and From Where? http://www.eia.gov/tools/faqs/faq.cfm?id=727&t=6.
3 The Goldman Sachs Group, Inc., The Ethane Advantage Has Evaporated: Is the SuperCycle Over? (December, 11, 2004), Exhibit 4, p.6.
4 The estimate of this elasticity is from an error correction model relating US employment and the cluster of oil jobs in Houston to local payroll job growth. Strong evidence was found for a long-term relationship to Houston total employment – working out over 6 to eight quarters -- for both the national economy and the oil cluster. Various versions of the model agreed closely on the elasticities cited here.
5 Perhaps because of the uncertainty about oil price, much of the discussion in the industry has not been about rigs counts or prices, but the percentage cuts in E&P as a result of low prices. For that reason, we used data on historical E&P expenditures, and related them to the number of basic jobs in energy. The statistical relationship was strong, showing that E&P spending has powerful short- and long-term effects on Houston’s energy employment. In the past, the analysis would have used the rig count in place of E&P, but rigs have become a less useful indicator of oil-field activity in recent years. Some industry analysts predict a decline of 500 rigs in 2015-2016. The downstream capital expenditures did not affect the number of jobs in the energy base, and was handled differently. This is discussed below.
6 There are probably only 100,000 or so white-collar jobs in the energy industry as we defined it. However, since each of these jobs support 3 or more nonbasic jobs, the cuts will reach well outside a narrowly defined energy base. Professionals in engineering, consulting, legal, accounting, personnel services, and many other categories will also find themselves as potential layoffs.
7 Total construction employment in Houston was estimated as a function of general economic trends and specific spending by the oil industry for downstream construction activity. A statistically significant relationship was found for both variables, again using an autoregressive error-correction model. Long-term estimates were similar in restricted and unrestricted estimates of differences, as well as in levels estimates.
8 For the energy base, the multiplier is even higher. We estimated above that a 1 percent (3000 job) increase in the energy base would result in 18,000 total jobs. Then the multiplier is (18,000 – 3,000)/3,000 = 5 nonbasic jobs per base worker.
Written by: Dr. Robert W. "Bill" Gilmer
Director, Institute for Regional Forecasting
December 16, 2014
Houston’s Outlook Brightens as Job Growth Picks up Strongly in Early 2014
Employment in Houston has surged over the first four months of the year, according to the payroll employment data released by the Texas Workforce Commission (TWC). Seasonally-adjusted data show 38,700 new jobs through April, or if the pace continued through the year 116,100 new jobs. If such a year materialized, it would compare to the monster addition of 115,000 new jobs in Houston’s metropolitan area in 2012, before 2013 saw job growth slow to 76,000.
The current data show a sharp reversal of trends that led to the 2013 slowdown, especially a dramatic turn-around in jobs tied to oil- and gas-related drilling. This raises several questions. First, are the data to be trusted? After all, TWC payroll figures are estimates based on a relatively thin sample and substantial statistical modeling; they will be subject to repeated revision in the months to come. Can we, for example, confirm the pick-up in growth in other data series? Second, if it is real, what is driving it? And finally, is it a temporary bump or likely to continue through the year or even beyond? The answers we offer below are that it is real; it is a mix of temporary and more persistent events; and that it provides an improved outlook for Houston in 2014 and 2015. Based on these results, we have raised our forecast for 2014 to 86,000 new jobs, an increase of 21,000.
Is It Real?
Figure 1 shows payroll job growth in Houston running at 4.5 percent annual rates through 2012, slowing to below 2.0 percent by the spring of 2013, and pulling back up to 4.5 percent in early 2014. Preliminary data show job growth slowing to 3.0 percent by April, but the cumulative total of 38,700 jobs added through April is much bigger than had been expected.
There are other data series that support the proposition that Houston’s economic growth accelerated early this year. The Houston Chapter of the Institute for Supply Management conducts a monthly survey of local purchasing managers that results in a purchasing managers’ index (PMI) similar to the manufacturing index released by the national organization on the first business day of each month. The PMI is set so that 50 is neutral, above 50 indicates expansion, and below 50 implies contraction. Following the end of the 2010 recession, Houston’s PMI was consistently reported as being 5-10 percentage points higher than the national index, largely reflecting the ability of oil- and natural gas-related activity to move Houston onto a stronger growth path. Last year, with oil and gas exploration pulling back in the wake of a collapse in natural gas prices, the Houston and national indexes moved to comparable levels, as seen in the left side of Figure 2. However, in early 2014, Houston again separates itself from the U.S. index. The PMI is comprised of seven individual series, including responses to questions about sales, production, lead times, and other factors. The right side of Figure 2 shows that, according to this data, sales and production led the early-year acceleration of local economic activity.
Figures on unemployment also indicate an acceleration of local growth. Unlike the payroll employment data, which is based on a survey of local firms, unemployment figures are compiled from a telephone survey of households, and provide another check on recent economic strength. Since the end of the recession in late 2009, the number of unemployed workers in Houston has been reduced by about one-third. Progress in reducing the number of unemployed slowed through 2013, and then accelerated in 2014. By April, the local unemployment rate fell to seasonally-adjusted 4.9 percent from 5.7 percent in December, to reach its lowest point since July 2008.
Where Does It Come From?
Given three independent sources of data that confirm an early-year growth spurt, we can have some confidence that it won’t all be revised away. The most likely suspects for causing such acceleration in Houston are always the goods-producing sectors – oil and natural gas, construction, and manufacturing. Figure 3 shows that oil- and gas-related employment had been running at 10 percent annual rates in late 2012, but had slowed to a mild contraction by late 2013. It then suddenly surges back to double-digit rates in recent months. Similarly, construction employment had stopped growing by mid-2013, only to surge back to double-digit growth early this year. Manufacturing remains flat, however, as did the manufacturing sectors most closely related to oil and gas exploration – fabricated metal and machinery.
A chart very similar to Figure 3 can be produced by just replacing the Houston data with that for the state of Texas. It shows statewide oil- and gas-related jobs spiking up to a 15 percent growth rate, and construction to near 6 percent. Weekly hours in Texas oil and gas have quickly risen from 42 to 47 hours. Houston is sharing (or perhaps leading) similar trends seen across the state.
Temporary? Persistent? Permanent?
Unfortunately, the data do not come with neat explanations attached, leaving us to speculate about what is happening here. Let’s start with oil and gas, and two likely possibilities.
- One possible cause of a jump like that seen in Figure 3 would be the relocation of a large number of jobs into the Houston area by a major producer like Chevron, Exxon, or Occidental, all of whom have announced that such relocations will occur. We have no specific knowledge, but such an early-year transition of a large number of jobs to Houston payrolls could explain part of the jump.
- Second, the cold winter of 2013-14 produced a very large cash bonus for natural gas producers, as increased heating demands pushed up depressed natural gas prices. A $1 increase in the price of natural gas per thousand cubic feet, if maintained for a month, would produce a $2 billion increase in revenues to producers. Based on the increased natural gas prices between November and March, this is an additional $10 billion or so that went into the pockets of natural gas producers.
In addition, natural gas futures prices remain above $4.50 through next spring, as prices are expected to remain high until we can re-establish normal inventories of natural gas. In the spring of 2014, natural gas inventories had been pulled down by the harsh winter to only 45 percent of normal. These continued high prices offer producers the opportunity to sell their production forward for additional revenues.
Bonus money delivered by the polar vortex is apparently being spent by natural gas producers – mostly on oil-directed drilling in the Permian Basin. While other land-based drilling in Texas remains flat, the Permian Basin’s rig count has surged from 450 to 550 working rigs since November. The Permian Basin is the greatest of U.S. oil fields, and this finally marks its transition to horizontal drilling from vertical drilling. It is rich in opportunities for horizontal drilling and fracturing, but has until now the basin has been so rich in other opportunities, opened up by high oil prices and technology, that it has been too busy to turn to shale plays. As always, much of the capital, the high level executive oversight, the engineering, and the equipment will come out of Houston.
This is a one-time, weather-related event, but with stimulus that should carry through 2014 and into early 2015. It could bring to Houston an unanticipated bonus of several thousand jobs.
The only clear fact that emerges from the recent data on construction is that about 40 percent of the increase is related to heavy construction for utilities, roads, lot development, and other infrastructure. Probably backward looking, this is a continued effort to catch up with the rapid growth that has shaped the local economy for more than a decade.
The rest of the construction growth in 2014 is general building construction – single-family residential, multi-family, retail, office, light industrial, and heavy industrial. Out of this group, most can also be categorized as backward-looking, catch-up activities in response to past growth. The exception is heavy construction on the east side of Houston, new petrochemical and liquefied natural gas (LNG) export facilities, some of which are now entering the construction phase.
Horizontal drilling and hydraulic fracturing of shale in the U.S. has produced an abundance of cheap natural gas and natural gas liquids. The liquids are produced along with the methane gas stream – butane, propane, ethane, and natural gasoline – and are used to produce intermediate products (petrochemicals) that ultimately become plastic or synthetic rubber. For example, ethane is used to make the chemical ethylene, the largest building block on the Ship Channel, which ultimately becomes polyethylene or polyvinyl chloride (PVC) plastic.
The abundance of natural gas liquids that have resulted from fraccing have resulted in low feedstock prices and major competitive advantages for U.S.-based petrochemical producers. Houston is home to a major share of the U.S. industry, with four of the largest eight ethylene complexes in the world located here, and it is about to be home to three more – two in Baytown and one in Freeport. Historically in the U.S., we have produced petrochemicals for the U.S. market and exported opportunistically. However, these new plants are being built for export to take advantage of low domestic feedstock prices.
Table 1 compares two large and very similar projects proposed by ChevronPhillips and Exxon, each with a cost of about $6 billion. To put the cost and scale of these figures in perspective, a 30-story office building downtown might cost $250 million. These projects are each the equivalent of 20-24 such towers.
Horizontal drilling and fracturing has also produced large new supplies of natural gas for export. Six projects to liquefy natural gas and export it have now been approved by the Department of Energy, with three of them in the Beaumont-Lake Charles area, and one in Freeport. Each plant carries a price tag of about $10 billion and an expected completion date of 2018. A number of other potential plants remain in the permitting phase.
All of these projects are anxious to move forward. The large petrochemical projects, in particular, are now moving through permitting, procuring long lead-time equipment, and are anxious to be the first to market in 2016 to take advantage of what are currently very lucrative profit margins. Each plant is also concerned about craft availability and machine shop capacity, and does not want to be the last in line for these services in a crowded field. It is likely that these multi-billion dollar projects are finally playing a significant role in Houston’s construction activity, and now that the surge has begun, they will continue to play a major role into 2018.
It is a strange witch’s brew of events that have come together to propel Houston’s economy through the first four months of the year: a possible one-time producer relocation, a weather-driven push in drilling, some more predictable catch-up activity in construction and lot development, and likely acceleration in long-anticipated petrochemical construction. Based on its timing and the apparent early-year growth we have already seen, we need to reconsider our forecast for 2014.
Our initial take on 2014 was that the slowdown in upstream hiring was to stay with us through 2014, and that downstream construction would accelerate slowly through the year. We forecast 65,000 jobs, down from 76,000 in 2013. This was to be a transition year, as drilling and exploration handed off responsibility for the local expansion to petrochemical construction. Growth would accelerate again in 2015 driven by growing heavy industrial construction. Based on the twice-over but temporary shocks to the upstream, plus downstream building activity that is accelerating earlier than expected, we have increased the forecast by 21,000 jobs to 86,000. The revision is based on less solid fact than we would like to have, and considerable uncertainty about how these various factors will play out over the next 18 to 24 months. But it is explicit recognition that the Houston economy is now growing quite strongly.
Written by: Dr. Robert W. "Bill" Gilmer
Director, Institute for Regional Forecasting
June 5, 2014
Benchmark Revisions Confirm Loss Of Momentum in Local Job Growth
The most important and current data we receive on Houston’s economy are the monthly payroll employment figures. The monthly data are estimates of the number of Houston workers eligible for unemployment insurance under the Employment Security Program that is operated by the State of Texas. The initial estimates are based on a sample of employers, but after several months the unemployment compensation data stabilize to provide an actual figure for the total number of covered employees, as well as the final number of workers in many key industries. Each March, the Bureau of Labor Statistics and Texas Workforce Commission look back to replace the sample data with actual — or at least much more accurate — counts, and to provide revised jobs numbers that cover a year or more. This latest benchmarking for Houston was released on March 7, and — as always — it rewrote recent economic history.
The latest revisions reached back as far as early 2012 and rebenchmarked data through the third quarter of 2013. Fourth quarter data and early 2014 remain sample estimates. The effect of the revision was to increase Houston’s estimated December to December job growth figure in 2012 from 105,700 to 114,500, and to reduce 2013 job growth from 82,000 to 76,200.
These new figures again confirm that 2012 was one of the greatest years ever for Houston job growth, but they also reinforce that we are seeing a broad pullback in the pace at which we are adding new jobs, and that slower growth began in mid-2013. Figure 1 shows annual rates of employment growth near 4.0 percent in 2012 and early 2013, but pulling back to a rate of 2.5 percent by year-end. The current 2013 estimate is a December to December change of 76,200 new jobs that closely matches our long-standing forecast of 74,000. The 2013 result is a 2.8 percent annual increase, well below the monster year of 2012, but also well above Houston’s long-term average of 2.2 percent.
Why the slowdown? A boom in upstream oil and gas drove domestic exploration and production spending at compound 20 percent annual rates from 2002-2012, but these E&P expenditures have been flat or growing much more slowly since the collapse in natural gas prices in late 2011. The Baker Hughes rig count is down by over 10 percent since its peak two years ago. The result for Houston-based oil and gas employment has been that the growth of jobs in oil production, oil services, fabricated metal, and machinery has shrunk from a 15 percent annual rate to nearer 2-3 percent. See Figure 2.
Economists also like to divide employment into basic and non-basic jobs. Basic jobs are those that drive growth by selling their products outside the 10-county metro area. The most obvious examples in Houston are energy related – oil services, oil production, machinery, refining, and petrochemicals. The non-basic jobs are followers that rely on base expansion for momentum, and are inherently local activities like laundries, dry cleaners, restaurants, and most construction and real estate. We don’t get rich taking in each other’s laundry or by building each other a new house or office building. Figure 3 shows the growth of base employment (proxied by mining and manufacturing) pulling back to rates near 2 or 3 percent, and the non-basic jobs are now falling into a similar trajectory. We know that over the long-run, basic and non-basic jobs will grow at the same rate.
In summary, the newly benchmarked revisions: 1) moved 8,800 jobs back to 2012 and, 2) showed 2013 job growth that was slower than previously thought, with 76,200 new jobs measured December to December. The data also show a rapid and sustained slowdown in basic employment led by upstream oil and gas. Our spring symposium on May 22 will be a search for a replacement for the momentum lost to this pullback in upstream E&P spending. We think the prospects for Houston remain solid, but we will rethink the current 2014 forecast of 64,000 jobs after looking closely at the petrochemical expansion underway on the east side of Houston, improved prospects for the U.S. economy, and a variety of other factors. Can one or more of these economic drivers step forward while Houston’s upstream sector steps back?
Written by Robert W. Gilmer, Ph.D.
Director, Institute for Regional Forecasting
March 10, 2014
Three Months of New Jobs Data for Houston Confirm Slowdown, Paint Slightly Better Picture for Economy
The shutdown of the federal government in early October delayed the release of September payroll employment and unemployment data for Houston until November 22, the regular release date for October figures. Counting the revision of August data that should have occurred in September, this latest news gives us insight into the last three months of local job growth. The new numbers confirm the slowdown already apparent in Houston’s job growth since mid-year, but also paint a slightly more optimistic picture of recent growth than earlier data.
Houston's payroll employment growth remains strong. (Figure 1) The prior data through August indicated that job growth had slowed on a year-to-date basis to an annualized 2.8 percent; the latest figures are 3.2 percent through October, a rate that would deliver well over 80,000 jobs if it continued through December. Both figures are excellent – well above Houston’s average of 2.2 percent annual growth since 1990 – and discussion of slower growth must be kept in this context. The slowing is from a beginning point of 4.0 percent job growth in 2012 and over 100,000 new payroll jobs.
There is good reason to be cautious about interpreting small changes in the data. First, strength has varied within the year, running four to six percent early in 2013, but averaging 2.7 percent during the last six months. Further, the most recent estimates from the Texas Workforce Commission have followed an erratic path, even when seasonally adjusted and smoothed with a three month average as in Figure 1. This suggests that the Commission may be struggling with the current estimates, and we could see significant revisions in coming months. For this reason, our forecast of local job growth remains near 75,000 for 2013.
The Houston unemployment rate tells much the same story. Figure 2 shows that Houston’s unemployment rate had been dropping faster than the U.S. since late 2010, but the trend was broken in 2012. Houston’s unemployment rate has been stagnant near 6 percent since last June, and the new data simply confirmed the pattern.
Why slower job growth? The primary cause is almost certainly the collapse of natural gas prices in late 2011, which led to 560 rigs being pulled away from natural gas directed drilling and a 2012 drop in the domestic rig count of over 12 percent. Figure 3 shows the rate of growth of jobs in Houston upstream sector (both producers and services) and in closely-related machinery and fabricated metal. Together, these sectors account for over 225,000 of the best-paid jobs for Houston. Both were growing at annualized rates of 15 to 20 percent in 2011, but have now settled in at about 5 percent. For the next 24 months a pick-up in this growth rate much above 5 percent seems unlikely.
In summary, the new data confirm a shift to slower but healthy growth in Houston. The data through October are slightly more optimistic about the extent of the slowdown, but the erratic month-to-month pattern of recent estimates raises questions about the durability of this interpretation. For now, we will stick with our forecast of 75,000 jobs for Houston in 2013 and 65,000 in 2014.
Written by: Dr. Robert W. “Bill” Gilmer
Director, Institute for Regional Forecasting
November 25, 2013
Houston Job Growth Slows at Mid-Year,
More Moderate Growth Ahead
Houston’s payroll job growth began to cool over the spring and summer, and has taken a step down from the rapid pace of 2012 and early 2013. Thanks to a very strong start to the year, year-to-date job growth through August is 2.8 percent; but the last six months have seen Houston’s growth slip to seasonally-adjusted annual rates near two percent. (Figure 1) The monthly employment changes reported by the Workforce Commission have swung widely, and the month-to-month revisions have been large, indicating difficulty in estimating recent job growth. The final revisions to the payroll employment data, which will come early next year, may be larger than usual.
Apart from the payroll data, a variety of other sources point to a cooling off of the Houston economy. The seasonally-adjusted unemployment rate has been flat near 6 percent all year; the Houston Purchasing Managers’ Index has fallen to levels similar to the U.S. in recent months; the weekly hours worked in Houston’s manufacturing sector were 49 hours per week a year ago, but 45 hours today. None of this is fatal to Houston’s economic outlook, but combined they point to a definite ratcheting down in the pace of Houston’s economic expansion.
The torrid pace of last year, 4 percent job growth and over 100,000 new payroll jobs, was probably not sustainable under the best circumstances, but certainly not when some of the fundamentals have moved against the local expansion. Houston’s economy depends on its ties to the U.S. economy, to oil and natural gas, and to global markets, and all three are less robust than a year ago.
The U.S. Outlook
The U.S. economy has provided little support to Houston since the end of the Great Recession. U.S. GDP growth has averaged only 2.2 percent, well below the long-term trend, and especially slow during a period of recovery from deep recession. Job growth has also disappointed, with current employment still 2 million jobs short of the prior peak in late 2007.
The good news is that the U.S. economy is finally showing signs of putting the financial crisis and recession behind it, perhaps ready to return to its long-term growth trend near 3 percent. The three areas of maximum damage sustained during the financial crisis were new home construction, personal consumption, and state and local government, and they have combined to slow growth since recovery began. The heart of the crisis was housing, and new single-family home construction fell by close to 80 percent. (Figure 2) The U.S. consumer was left seriously over-burdened by debt, much of it mortgage debt, and consumption suffered as they worked down this debt load. State and local governments were hurt by the decline in sales tax revenues as consumption fell, and by the loss of property taxes, and home values declined.
All three sectors are nearing the point where we can declare the financial crisis over and the damage finally repaired. The U.S. consumer, for example, has worked off much of the debt load incurred during the housing boom, and debt-to-income ratios are rapidly moving toward pre-crisis levels. This offers the opportunity for much freer spending on housing, autos, and general retail.
Also, the housing market made a definitive turn upward in the summer of 2012, with home prices rising across the country. Along with these price increases, the prospect of higher interest rates pushed many reluctant home buyers off the fence, and sales of new and existing homes began to rise quickly. New single-family permits are still only running at annual rates of 600,000, well short of the 1.0–1.2 million that would be considered normal, but new home construction is rising and adding to GDP again.
Finally, state and local government revenues were hurt twice over. The decline in consumption slowed sales tax collection; then the decline in home prices hurt property tax revenues. A combination of budget cuts and rising revenues have now returned most states to a balanced budget, while the prospect of improved consumer spending and rising property values can only enhance the outlook.
With crisis-related damage healed, most forecasts for U.S. economic growth have turned quite optimistic, with expectations growing for a return soon to historic long-term growth rates of 3.0 percent or higher. The bad news is that we are not quite there yet. Stronger growth has been delayed by $200 billion dollars in new federal tax revenues from payrolls, to fund the Affordable Care Act, and from expired tax cuts. Add to this fiscal drag another $85 billion in federal spending cuts. Real GDP growth in 2013 is again forecast to be 2 percent or less, adding yet another year of muddling through for the U.S. economy. But a variety of forecasters ranging from the Survey of Professional Forecasters to Goldman Sachs now see strong growth in 2014 and beyond, rising to a steady 3 percent GDP growth and staying there for the long-term.
Oil and Natural Gas
For the last decade, Houston has found itself at the heart of the greatest oil and natural gas boom in U.S. economic history, led primarily by drilling for oil and gas in shale. Capital expenditures for exploration and production (E&P) reached $279 billion in 2012, increasing by a factor of six in a decade. Even allowing for inflation, the 2012 expenditures were 4 times those of 1982, the point at which the great oil bust of the 1980’s began. (Figure 3) These E&P expenditures peaked in 2011, however, they were down about 10 percent in 2012, and are expected to be flat in 2013 compared to last year. The Baker Hughes rig count – a short-term indicator of the drilling cycle – tells much the same story as the expenditure data. The collapse in natural gas prices in 2011 led to a 13 percent decline in the rig count in 2012, and it has been flat through 2013. Drilling activity continues at a very high level, but is no longer growing, pushing the boundaries of the Houston economy and stimulating growth.
The outlook for a near-term pick-up in E&P spending is not strong. Our forecast is for very modest increases in the rig count through 2014, picking up in 2015 and beyond. These increases will be driven by stronger demand for natural gas due to faster U.S. growth, exports of liquefied natural gas, and the start-up of a series of petrochemical plants built to capitalize on low prices for natural gas liquids.
There is a built-in cushion against this upstream slowdown. Oil-directed drilling remains very profitable and stable; there is much midstream pipeline and processing infrastructure necessary to connect new shale plays to the Gulf Coast; and low natural gas prices have kicked off a boom in downstream petrochemical construction. The major risk to the near-term future would be a decline in the price of oil, with 80 percent of drilling – both domestic and abroad – now directed to oil.
The Global Economy
For the last decade, the global economy has been driven by China, India, Brazil, and other emerging markets. Houston has achieved a high growth rate through its ability to reach past a sluggish U.S. economy and to tap into the growth of these developing nations. This was done in part by exporting local products – crude oil, refined products, machinery, and petrochemicals – to these high-growth countries. In 2012, Houston passed New York to become the number one merchandise exporter among U.S. metropolitan areas. Perhaps more important, the growth of these emerging economies was responsible for very large increases in the price of oil, metals, agricultural products and food. Oil rose further and faster than other commodities, with all the increases in global oil demand over the last 10 years stemming from emerging markets. Houston has been a substantial beneficiary of these high oil prices.
Recent growth in Brazil, China, and India has slowed. (Table 1) After a decade of double-digit increases, China’s growth in 2012 fell to 7.8 percent, as the Chinese attempt a risky transition from an export-led economy to one driven by domestic demand. India saw growth slow from 7.7 percent in 2011 to 4.0 percent in 2012, with few signs of a pick-up in 2013. Brazil’s growth in 2012 fell to less than one percent.
This emerging-market slowdown is one more of the fundamentals of the Houston economy that is less robust than a year ago. And they pose risks going forward, perhaps now more than usual because of the slower pace of growth. Certainly some event, like the financial crisis that swept through emerging markets in 1997, would collapse the price of oil and threaten Houston’s near-term growth. Over the longer-term, the greatest risk would be the inability of emerging markets to sustain the high growth rates of recent years. There is a long history of developing nations that grow strongly, reach a mid-life crisis, and then require major legal, financial, energy, or labor reforms to continue their strong growth. Unable to make these reforms, they simply slow to a pedestrian pace. For Houston, it would mean the loss of exports and perhaps the loss of the major prop for the price of oil.
Outlook for Houston
The economic backdrop for Houston is changing. Energy and global expansion have stepped back as sources of growth, even as they continue to operate at very high levels. Meanwhile, as early as next year the U.S. economy should take a much bigger role in Houston’s economic expansion. Table 2 summarizes our current outlook for Houston through 2017, with local job growth falling to 73,700 this year, and 65,100 in 2014. The base case depends on a return to solid growth in the U.S. economy and a growing rig count by 2014. The conservative forecast assumes a less robust turn upward in U.S. GDP, with growth remaining near 2 percent.
Although the profile of the current slowdown in local economic activity is becoming clear, the future outlined in Table 2 is growing murky again. History tells us that the official data typically are the least reliable when we need them most – at moments of change like this one. We are likely to see major revisions to the employment data for this year. The continued debt ceiling debates in Washington, the ongoing sequester of federal funds, and Federal Reserve monetary policy will also have an important bearing on near-term growth prospects for both the U.S. and for Houston. For now the forecast for Houston remains much like that made last spring, but it is an uncertain and fluid time that bears close watching.
Written by: Dr. Robert W. “Bill” Gilmer
Director, Institute for Regional Forecasting
September 27, 2013