Note on GDP revisions and changing the narrative

Yesterday BEA released the “third” estimate of the GDP for the first (Q1) and second (Q2) quarters of 2023. As suspected, there was some change in the overall economic growth rate from the earlier “advance” and “second” estimates. But you had to look closely to find it. And most analysts seem to treat this as old news. The revisions, however, remind us that much of our economic data is preliminary. Statistical agencies are constantly trying to maintain a balance between timely but incomplete data and more accurate data delivered in a less timely fashion. Hence the importance of paying attention to the revisions.

The revised GDP data shows that the economy grew by more in Q1 2023 than originally indicated and less in Q2. The revision changes the narrative slightly. The size of GDP growth slowed down between Q1 and Q2, rather than increasing.

With respect to knowledge-based investments, software and information processing equipment also grew by more in Q1 2023 than originally indicated and less in Q2. In fact, investment in information processing equipment actually decline in Q2 2023, whereas the advanced estimates showed a slight growth in Q2. There was only a slight revision to the R&D expenditures in Q1 2023, but a large difference in Q2 estimates with the third estimates showing almost no growth in Q2. Expenditures for Entertainment, literary, and artistic originals bucked the trend, showing less growth in Q1 and more growth in Q2.

The revisions are largely due to a major change in the calculation of real GDP. The advanced estimates use 2012 data as its baseline for converting current-dollar data to “real” (inflation adjusted) data. This chain-dollar data has been updated to use 2017 as the baseline. There were also updates of current-dollar data for 2013 to 2023. See https://www.bea.gov/information-updates-national-economic-accounts.

The source of the revisions done not change my point. Economic data is subject to revisions which can be large enough to change the narrative. Case in point was the revisions to the February and March employment data and the follow up revisions to the March and April data. The new data showed growth in February and March was significantly weaker that earlier estimated. The next month’s revision showed an economy that was stronger than previously reported.

I continue to wonder if the Federal Reserve’s actions may have been different had they been looking at the updated employment data earlier this year.

[Note: I define knowledge-related investment as the combination of investment in Information Processing Equipment, R&D, and Software. The first of these three categories is reported in the GDP data as a subcategory of Non-residential Fixed Investment: Equipment. The latter two are reported as subcategories of Non-residential Fixed Investment: Intellectual Property Products.]

July’s US trade deficit grew as intangibles surplus declines

The Bureau of Economic Analysis (BEA) reported this morning that the trade deficit grew by $1.3 billion in July to $65.0 billion as imports grew more than exports. The good news part of July’s data is that the increase in exports was a reversal of the recent trend of declines over most of this year. The other good news is that trade in tangible services finally posted a surplus after being hit hard by the pandemic and recording about 3 years of monthly deficits. The not-good news is that the surplus in intangibles was slightly lower in July as imports increased more than exports.

The lower surplus was largely due to an increase in imports of Business Services combined with a slowdown in exports, which was enough to result in a rare monthly decline. Lower surpluses in Financial Services and in Telecommunications, Computer & Information Services also contributed to the lower overall intangibles’ surplus, as did a higher deficit in Personal, Cultural, and Recreational Services. The surplus in Charges for the Use of Intellectual Property grew, but only because of a drop in imports. Maintenance & Repair Services surplus rose slightly as exports grew and imports declined.

As I noted in last month’s posting, four sectors account for most of the overall trade balance. July’s data, including revision for the first half of 2023, shows different trend lines from the ones I described last month. Rather than showing steady growth, trade surplus in Financial Services has stagnated after a big increase in January and February and a big drop in April. Also, contrary to last month’s discussion, revised data indicates that the deficit in Insurance Services is improving somewhat in 5 of the last 7 months with big improvements in January and February. Exports have generally increased as imports have remained basically flat after a big decline in January. The surplus in Business Services generally continues to grow but can vary dramatically from month to month. Charges for the Use of Intellectual Property has also shown wide monthly variations but has generally been flat for the past decade.

Over the long-term combined effect of these sectors has been a positive overall trend line, with growth continuing since the recovery from the pandemic, due largely to Business Services. However, with the volatility of the Business Services sector and the slow to no improvements in the other three largest sectors of intangibles, there seems to be little reason to expect a major improvement in the U.S. trade deficit. And as the chart below shows, the deficit in goods trade overwhelms any surplus in intangible services, even if combined with tangible services.

Note: I define tangible services as consisting of the BEA’s categories of Transport, Travel, Construction, and Government Goods & Services n.i.e.

Local DC area governments and the hybrid model of work organization

Last week, the Metropolitan Washington Council of Governments (COG) weighted in on the Fed’s return-to-office (RTO) efforts (or lack thereof). In a letter to the Office of Management and Budget (OMB), the COG joined the push by many in the Washington area, including D.C. Mayor Muriel Bowser, for a greater in-person presence by Federal workers. (The letter is available here and the Washington Post story here.)

The letter was, however, differed from other such appeals. Rather than coming from elected official or business leaders, it was signed by the Chief Administrative Officers of the various local municipalities—the people who have had to deal with the realities of managing an in-person and remote workforce. And the letter was not simply a call for OMB to do something. It was an offer to work together with the Feds and share “lessons learned from our collective experience, as our local governments have transitioned over the last several years from a remote environment to in-person and hybrid schedules.”

And it was an explicit endorsement of the hybrid model, describing their experiences:

For those who are eligible to telework, employees typically report to work in-person two to three days a week (not including weekends). We have found that this strikes an appropriate balance and provides the best level of service for taxpayers. Being able to work together, troubleshoot problems, take on big ideas, and provide face-to-face service for our residents is achieved while still providing flexibility for our personnel to work from home.

I would hope that the Feds will take the COG’s offer seriously. But I would urge COG to move ahead with its own analysis regardless. There is great value in the COG taking steps to document those lessons learned in terms of implementing a hybrid model. One area of lessons learned could focus on understanding the various alternatives and key decision points. Another set of lessons learning could involve how to interact with various stakeholder groups, including working with the federal government employee unions.

And, there is important information to be gained about the implications of the adoption of a hybrid model. One question concerns the impact on space requirements of transitioning from full time remote work to a hybrid model.

We know that the increase in remote worker during the pandemic greatly reduced the need for office space, which led to a dramatic rise in vacancy rates. It has also led to increased attention to the issue of underutilized office space (see GAO report and Washington Post story). It is unclear whether adoption of a hybrid model will allow government agencies to reduce office space. The repurposing of excess government office space is seen as a key element in strategies for revitalizing downtown.

As I discussed in an earlier posting, the hybrid model is based on the (perceived) need to facilitate various forms of personal interactions. Such interactions (worker coordination, sharing of tacit knowledge, mentoring, and learning by example/learning by doing) are generally believed to require in-person contact. Which requires people to be in the office at the same time on the same days. Thus, the work pattern (for a 3 days in-person; 2 days remote model) is high levels of workers in the office on Tuesday, Wednesday, and Thursday with a low number of in-person workers on Monday and Friday.

Such as work pattern requires enough space for everyone (or at least the projected “full load” of in-person Tues-Thurs workers). Having enough space for the Tues-Thurs crowd will likely be lower that the pre-pandemic level (when there was only a small level of completely remote work) but greater than today’s level because of the need to provide space for those workers shifting from full time remote to coming in to the office for part of the week.

An analysis of the experience of the COG members with implementing a hybrid model and dealing with office space issue would be very helpful. Specifically, has the adoption of the hybrid model led to a continued lower level of office space requirements (which would allow for repurposing of the excess)? Or has the hybrid model resulted in the need to maintain a level of office space to accommodate the 2 or 3 days a week when everyone is supposed to be in the office (leading to underutilized space, but not vacant space).

The second area of a COG analysis is in the economic impact. In the earlier posting, I also speculated that the adoption of a hybrid model would not necessarily be enough to revitalize downtown. The Tues-Thurs model replaces only portion of the pre-pandemic’s level of economic activity. Is a 3 day a week rush hour enough to restore transit-systems’ financing? Is it enough foot traffic to support all of the secondary businesses—the restaurants, bars, coffee shops, dry cleaners and parking lots that depend upon on office commuters.

Already downtown economic activity resembles a long weekend. For example, Friday’s have changed from a get-away day of long lunches and early departures to a stay-away day. The economic fallout has been noticeable. As the Washington Post reports:

Those shifting norms are rippling across the economy and reshaping business patterns for commercial real estate firms, parking garage operators and the many eateries that cater to workers during the week. The drop-off in office work, particularly on Fridays, has led coffee shops to reduce their hours, delis to rethink staffing and bars like Pat’s Tap in Minneapolis to kick off happy hour earlier than ever — starting at 2 p.m.

“Since they’re not at the office, people come in early to pluck away at their laptops while they sip a cocktail or two,” said General Manager Dave Robinson. “By 4:30 or 5 on Fridays, we’re completely full.”

But lunchtime haunts that once saw large crowds on Fridays say they’re struggling. The drop-off has been particularly stark at Manny’s Cafeteria & Delicatessen in Chicago. Business on Fridays is down 30 percent from pre-pandemic levels.

“It’s painful,” owner Dan Raskin said. “Before the pandemic, Friday was the busiest day of the week — people would have an easier day at work and go out with their friends for lunch — but now it’s one of the slowest.”

That’s also the case at LAZ Parking, which operates more than 3,000 garages nationwide. Demand on Mondays and Fridays is much lower — by about 20 percent — than it is midweek, said Leo Villafana, the company’s vice president for the Mid-Atlantic region. Wednesdays are the busiest days, though even when people do come in, they tend to stay for shorter periods.

It is unclear to me whether the adoption of a 3/2 hybrid model will help alleviate this “long weekend” problem.

—

It is often said that the states are laboratories of democracy. The same can be said of cities and towns. Case in point: the offer by the Metropolitan Washington Council of Governments (COG) to share their experiences and lessons learned on managing the issue of remote work with the Federal government.

There are at least three areas where the Federal government can learn from COG member’s experiences:

1. Understanding and managing/implementing the difference types of hybrids
2. Whether and how implement a hybrid model can free up office space
3. To what extent can secondary business (restaurants, bars, coffee shops, dry cleaners, parking lots, etc.) be helped by implementing a hybrid model

Governments need a deeper understanding in all three areas. I hope the COG and the Feds can get together on this. At a minimum however, COG should go ahead on its own to capture the lessons learned. Sharing information among the COG members is a worthy activity in and of itself. Participation by the Feds in this activity would bring additional benefits to all involve.

Labor market continues to grow, but reliant on one intangible-producing industry: Educational & Health Services

The U.S. economy continues to chug along. Nonfarm payrolls grew by 187,000 jobs in August, according to the Bureau of Labor Statistics. But the growth was not as fast as previously reported. Employment for June was revised down by 80,000 to 105,000 and July was revised down by 30,000 to an increase of 157,000. As the BLS notes, employment in June and July combined is 110,000 lower than previously reported.

Employment growth in intangible-producing industries outpaced tangible-producing industries. In August, payrolls in intangible-producing industries were up by 111,700 jobs while payrolls in tangible-producing industries increased by 75,700—unlike July where the increases in payrolls in intangible-producing industries and tangible-producing industries were roughly the same. Once again, the growth was led by Educational & Health Services (excluding tangible services) which grew by 82,200 jobs in August after growing by 77,600 jobs (revised) in July. And employment in Tangible education and health services was up by 19,800.

Employment in Professional and Business Services (excluding tangible services) grew by 10,000 jobs. This reversed the decline in June and July. However, the August increase was much smaller than in the first half of 2023 where this sector had been a leading source of new jobs. Also of note, Information (excluding telecommunications) employment dropped by 11,400, continuing 4 months of decline. Arts, Entertainment & Recreation payrolls grew by a healthy 16,700. Government (excluding Postal Service) added only 5,200 jobs—a dramatic slowdown in what was a major source of employment growth.

With respect to tangible-producing industries, Construction and Mining was up 20,000. Accommodation and Food Services saw an increase of 23,500 jobs. Employment in Manufacturing was up by 16,000 after having been relatively weak for all of this year.

Bottom line is that Educational & Health Services is driving the US labor market, providing almost 40% of this year’s growth in employment.

Note that the industry categories of Educational & Health Services (excluding tangible services) and Professional and Business Services (excluding tangible services) are variations of the BLS categories with jobs that are more physical move to new categories under tangible services. For example, Professional & Business Services are divided into tangible and intangible-producing. Waste Management & Remediation Services and Services to Buildings & Dwellings are broken out of the major industry group and assigned to a new category of Tangible Business Services. For more on the categories, see my explanation of the methodology in an earlier posting.