Winter 2020-2021 edition of our 22-page Forecast Starts report, U.S. & Canada. ..

Article source: ConstructConnect

Highlights

•  US construction starts fell 28.2% year-on-year and 19.9% year-to-date in Q3 2020. Non-residential building experienced the steepest decline, as the pandemic has weighed on the prospects of sectors such as hospitality, retail, and office working. The decline in civil engineering construction was exacerbated by some mega projects in 2019 falling out of the calculation. The decline in residential homebuilding has been driven by construction of apartments, with the single-family segment holding steady.
• The pace of the recovery has slowed markedly, driven by the expiry of fiscal stimulus, rising infection rates, and election uncertainty. Our baseline forecast assumes that Congress will eventually pass a new fiscal aid package, but even if Biden wins the election and delivers fiscal aid in 2021, the economy would stall in Q4 2020. We see the economy contracting 3.5% in 2020 and expect a relatively soft rebound of 3.7% in 2021.
•  US construction activity is forecast to decline more deeply in 2020 than during the global financial crisis of 2008-09. Non-residential construction will see the sharpest falls amid weak business investment. Residential activity will hold up relatively well, particularly within single-family construction. Looking ahead, we forecast a rebound in growth from 2021 but it will be slow and unevenly distributed.
• Total construction starts in Canada in Q3 declined 43.9% year-on-year and 28.6% year-to-date, with declines in all three headline segments. Restricted construction earlier in the year has largely resumed, but strong activity in Q3 2019 has fallen out of the annual calculation.
• Headline construction starts are forecast to decline by nearly 30% in 2020, following a contraction last year as well. In the recovery phase, new construction activity is expected to grow at a double-digit pace through 2023, but it is not expected to regain its pre-pandemic high (in 2018) until 2023.

Overview

Further declines in US construction in Q3

Total US construction starts fell steeply in Q3, shrinking 28.2% from a year earlier and 19.9% year-to-date (from January to September compared to the same period in the previous year). Non-residential building experienced the largest decline, down 46.3% year-on-year (y/y) and 33.7% year-to-date (ytd). Engineering construction declined 29.2% y/y and 15.9% ytd. New residen- tial construction experienced the smallest decline, shrinking 5.9% y/y and 6% ytd.

One feature of construction in 2020 rela- tive to that in 2019 has been fewer mega projects valued at over $1 billion. 2019 was an exceptional year for mega projects with construction starting on 35 such projects worth a total of $79.1 billion. August 2019 was the single biggest month for mega proj- ects, with groundbreaking on five projects totalling $18.7 billion. Mega projects falling out of the annual calculation is certainly exacerbating both the y/y and ytd declines in 2020, but underlying construction activity is also weak, with the whole-year value of construction expected to be at its lowest level since 2015.

Non-residential building has fallen from January to September relative to the same period in 2019 in almost all segments. Groundbreaking on hotels, sports facili- ties & convention centers, transportation infrastructure, and factories all posted ytd declines exceeding 50%. Factory construc- tion saw some new mega projects started

* In 2020, boosted by stimulus checks and unemployment insurance top-ups.

in Q3, including a $2.3 billion GM battery plant in Ohio, linked to electric vehicle production, and a $1.1 billion Tesla plant in Texas, but a new $10 billion ethane cracker plant started last year in Texas fell out of the calculation. The weakness in hotel construction is hardly surprising, given the impact that the pandemic has had on hospitality. Private office building has also been notably weak so far in 2020, down 36.8% ytd, as the pandemic has accelerated a trend towards home working. Perhaps more surprisingly, new hospital construc- tion has also been anemic so far in 2020, falling 45.7% ytd. Within non-residential building, positive ytd growth has only been seen in libraries & museums, courthouses, police & fire, prisons, and military (boosted by groundbreaking on a $1.75 billion project earlier in the year).

In the engineering sector, only new water & sewage treatment facilities posted posi- tive ytd growth in starts, although new con- struction of roads and dams, canals & marine projects were limited to single-digit ytd declines. The largest contraction has been in construction of new power infrastructure,

down nearly 77% ytd (and nearly 90% y/y), but there were three mega projects in Q3 2019 which have dropped out of the calcu- lation. The miscellaneous civil sector, which includes oil & gas, pipelines, rail, and tunnel construction projects, declined more than 25% ytd and more than 50% y/y, as a couple of rail projects in Q3 2019 fell out of the annual calculation. This was partially offset by a $1.4 billion light rail expansion started in California in July.

Residential construction has been the most robust so far in 2020, driven entirely by growth in the single-family segment. Groundbreaking on new single-family homes rose 2.8% ytd and 11.4% y/y, as demand for additional space has grown as more people work from home. Low interest rates and steady wage growth among white collar workers have also helped underpin this trend. By contrast, new multi-family construction decreased 24.9% ytd and 41.7% y/y.

Other construction indicators have been slightly better than the starts data. Put-in- place construction from the Census Bureau, for example, was 2.5% higher than a year

earlier in August. However, this is a measure of continued construction activity, including the mega projects started last year, so it could be some time before the recent weak- ness in starts is fully realized in the put-in- place figure. After shedding over 1 million jobs in April and March, the initial employ- ment recovery in the construction sector was rapid, with 456,000 jobs and 159,000 jobs added in May and June respectively. Job growth has since slowed, with monthly employment gains rising by an average of 25,000 in Q3. Total construction employ- ment in September was 5.2% below its level in February, slightly ahead of the 7% decline for the economy as a whole.

An economy at a crossroads as we approach 2021

The pace of the US recovery has slowed markedly. The economy has regained more than two thirds of the output lost in Q2, but with fiscal stimulus expiring, flu season approaching and election uncertainty rising, the main question is how strong the economy will be going into Q4.

Forward-looking indicators are increas- ingly suggestive of a recovery at risk of stalling. While retail sales had a strong Sep- tember, rising 1.9% on the month, high-fre- quency data suggest the consumer recovery could struggle to maintain momentum in Q4, and sentiment data is still 20 points below pre-crisis levels.

In addition, the September jobs report showed nonfarm payrolls rising just 661,000, the smallest increase in five months. Employment remains a staggering 10.7mn below pre-Covid levels while the unemploy- ment rate of 7.9% is still close to previous recession peaks.

Business activity has also shown the first real signs that the recovery is losing momentum. Industrial production was down 0.6% in September as manufacturing fell for the first time in five months. Durable goods orders also showed weakness, advancing a tepid 0.4%

We believe Congress will eventually pass a fiscal aid package worth around $1tn, including transfers to households. However, given the heightened political volatility and the current administration’s recent opposi- tion to a comprehensive stimulus package, we fear a large downside risk to growth. Even if Biden wins the election and delivers fiscal aid in 2021, the economic recovery would stall in Q4 2020 with annualized q/q growth decelerating to about 1.0%.

Fed Chair Powell stressed the importance of conditional dovish forward guidance in the September FOMC statement. The Fed will keep interest rates at the effective lower bound until the economy is assessed to be at full employment, and inflation has risen to 2% and is likely to moderately exceed that level for some time. We expect the Fed will not raise interest rates until mid-2024 at the earliest. Lending from the Fed’s emer- gency facilities has remained modest, total- ling just under $95 billion. We expect the Fed’s balance sheet to be around $8 trillion by year-end.

Even assuming fresh stimulus, we see the economy contracting 3.5% in 2020 and expect a relatively soft GDP rebound of around 3.7% in 2021.

Construction has reached its nadir, but record annual drop forecast in 2020

Total US construction starts are fore- cast to decline by 18.8% this year as the US economy is buffeted by the Covid-19 pan- demic and associated voluntary and man- dated restrictions on economic activity. The non-residential segment is set to be the worst hit as private business investment has been scaled back significantly in the wake of the crisis. Residential starts are forecast

to fall by 6.7% this year, the smallest decline of the headline subsectors. The residential sector, particularly single-family, has held up well during the lockdown and demand should ensure stable growth going forward. Civil engineering is forecast to decline by 15.7% this year, but with a robust rebound forecast for 2021.

The decline in the residential sector is entirely driven by the multi-family segment, which is forecast to decline by 26%. By contrast, the single-family segment is fore- cast to grow this year, expanding by 2.8% despite the extraordinary conditions facing the sector. This relatively good performance is due to a number of factors. The most important is that social distancing is less of an obstacle to single-family construction than for other settings. Much of the con- struction takes place outdoors, and workers can effectively maintain safe distance in line with government guidance. Furthermore, a significant portion of single-family con- struction takes place in suburban and rural settings, away from built-up cities where pandemic restrictions can often be more restrictive.

Secondly, the underlying fundamentals supporting robust single-family growth remain in place. Pent-up demand for single- family housing among millennials remains a strong driving factor for new homebuilding. One of the emerging trends to come out of the Covid-19 pandemic is the move towards remote working. This will tend to reduce the demand for central city living versus suburban living, which should support single-family construction at the expense of multi-family construction. It remains to be seen how persistent this trend will be, but we do see it adding further weight to our robust forecast for single-family con- struction.

The non-residential sector by contrast remains lackluster. A sharp drop in business investment has led to a collapse in demand across commercial and industrial construc- tion. Office starts are forecast to decline by 43% this year with only a 14% rebound in 2021. The outlook for office construction going forward is one of the most uncer- tain parts of all non-residential. The move toward working from home will undoubt- edly reduce demand from firms to maintain office space for their workers. The pandemic has demonstrated that many office workers can work just as efficiently from home as from a central location, a point that also reduces firm’s fixed costs — thereby further incentivising work-from-home becoming a more long-term solution. Private office building is not forecast to regain its pre- pandemic level by the end of our forecast period in 2024.

The retail sector is also set to see sharp declines this year with only a slight rebound in 2021. Bricks-and-mortar retail stores have long been on a downward slide as more cost-efficient e-retail businesses take an ever-larger slice of consumer spending. The Covid-19 pandemic has exacerbated this trend as people have been forced to shop online — retail e-commerce sales rocketed 32% in Q2 2020, compared to a 4% drop in overall retail sales. This sudden shift has led to some longstanding bricks- and-mortar retailers going out of business permanently.

Starts in the manufacturing sector are set to pull back 65% this year, though it should be noted this is partly due to several mega projects from 2019 falling out the cal- culation. Indeed, manufacturing starts can be quite volatile due to such large-scale projects. Nonetheless demand for manu- factured products has also declined as a result of the pandemic and subsequent lockdown leading to reduced demand for construction services in the sector. Looking ahead, the picture is more positive as the re-shoring of factories to the US con- tinues (likely boosted by the pandemic), low energy prices and a robust domestic market drive demand for construction of new manufacturing space.

Civil engineering is forecast to fall by 15.7% this year. The most significant decline stems from the miscellaneous category, including projects in the oil and gas sector, which is forecast to fall by 31%. Relatively weak oil and gas prices have led to many wells cutting operations and investment for new infrastructure in the sector is being pared back. We forecast a mediocre rebound next year, but a continued low oil price environment will keep a lid on poten- tial investment into oil & gas projects.

Civil engineering has generally been less impacted by the pandemic than non-res- idential because of its heavy dependence on public funding. Whereas non-residential investment is typically determined by pri- vate investment levels, civil engineering projects are usually government funded and as such, less swayed by shocks to the economy — and to the extent that govern- ment infrastructure spending is called upon to counter economic weakness, can some- times be countercyclical. This is certainly true of road construction and wastewater & sewage construction, both of which are forecast to outperform this year. Looking forward, there remains a significant diver- gence between infrastructure need and actual funding assigned to meet that need. Much of US infrastructure is in a poor state and has been for some time, but the polit- ical will needed to push through significant reforms remains elusive.

Steep declines in Canadian construction

Total construction starts in Canada declines 43.9% y/y in Q3 and 28.6% ytd with pullbacks in all three headline seg- ments. Restricted construction activity in Quebec and Ontario earlier in the year has weighed on the ytd calculation, but construction had largely restarted by mid- May, so it does not explain the steep y/y fall, which was instead the result of strong activity in Q3 2019.

From a sectoral perspective, non-res- idential building suffered the steepest decline, down 63.5% y/y and 31.6% ytd. All non-residential segments except for police & fire declined in y/y terms, and in ytd terms, only the industrial, warehousing, and trans- portation segments were up, thanks to large projects earlier in the year. Civil engineering construction fell 37.5% y/y and 24.7% ytd. All sectors except for dams, canals & marine work and power infrastructure experienced y/y declines. In ytd terms, all sectors con- tracted except for bridges, which was boosted by a project in British Columbia earlier in the year. In August, work began on the $3.5 billion Keystone XL pipeline in Alberta, included in the miscellaneous civil sector, but a large project in British Columbia in Q3 of last year meant the sector starts contracted. Residential construction posted the smallest decrease, down 33.1% y/y and 30% ytd. Like in the US, construc- tion of new single-family homes has been more resilient, as demand has grown for a larger living space, but multi-family home building has fallen off a cliff.

GDP plunged at an historic pace of 11.5% quarter-on-quarter in Q2, and monthly trends point to a strong 10.3% rebound in Q3 as lockdown restrictions eased, albeit only partially recouping the output lost in H1. With Covid-19 cases once again on the rise and a retightening of lockdown restric- tions underway in parts of the country, all the signs point to a much weaker advance in Q4 2020. Canada’s fiscal response to Covid-19 has now grown to C$551bn (23.9% of GDP), and the federal govern- ment has recommitted to do “whatever it takes” to support the economy through the pandemic. The Bank of Canada has also reiterated its commitment to provide extraordinary monetary policy support. We expect the Bank will hold the overnight rate at the effective lower bound until early 2024 while continuing to grow its quanti- tative easing program. Overall, we project GDP will fall 5.3% in 2020, and rise 5.9% in 2021.

Headline construction starts are fore- cast to decline by 28.3% in 2020, following a 13.6% contraction in 2019. New construc- tion activity is expected to rebound by 27.1% in 2021 and grow at a double-digit pace through 2023, but it is not expected to regain its pre-pandemic high (in 2018) until 2023.

Residential starts are expected to decline 31.3% this year, before rebounding to 18.3% growth in 2021. New single- family construction is expected to be little changed this year (shrinking just 0.3%), with construction of new apartments sub- siding by more than 50%. Growth during the recovery period is expected to be stronger in the multi-family segment than in the single-family segment. However, this is largely due to base effects given the extent of the decline in 2020 — multi-family home construction is still expected to be lower than its pre-pandemic level at the end of our forecast period in 2024. As in the US, demographic factors tend to favor single- family construction, due to increasing household formation within the millennial generation, and the pandemic has only accelerated the trend away from city living. Indeed, single-family home construction is expected to regain its 2019 level next year, although after declines since 2018, it does not return to its 2017 peak until the end of the forecast period in 2024.

New non-residential building is expected to decline 33.1% in 2020. All sub-sectors are expected to decline this year, except for transportation terminals, boosted by a large project in Ottawa earlier in the year, and warehouses, boosted by strong investment in logistics capabilities. Both of these sectors are also likely to grow in the recovery phase. Several new projects are planned in the transportation sector, while structural shifts towards online retail should see continued demand for new warehousing space. However, some vacant retail space may be able to be repurposed to use as warehousing, so growth in the latter category is expected to slow from its pace this year. By contrast, the pandemic has led to structural shifts in a number of sectors due to social distancing, travel restrictions, and increased home working. Sectors which are not expected to regain their pre-pandemic highs include amuse- ment, shopping & retail, hotels, private offices, and parking garages.

Engineering construction is forecast to shrink 22% in 2020 with declines in all sectors except bridges. However, during the recovery period, the sector should be the fastest to return to its pre-pandemic high, with activity expected to exceed its 2019 level by 2022. Public funding for infrastructure projects such as roads, rail- ways, bridges and other crossings should be forthcoming. Large projects are also planned in the power sector and in the oil & gas supply chain which, if they go ahead, will give a large boost to engi- neering construction.

EXPLANATION: Table 3 conforms to the type-of-structure ordering adopted by many firms and organizations in the industry. Specifically, it breaks non-residential building into ICI work (i.e., industrial, commercial and institutional), since each has its own set of economic and demographic drivers.

Table 4 presents an alternative, perhaps more user-friendly and intuitive, type-of-structure ordering that matches how the data appears in ConstructConnect’s on-line product ‘Insight’.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect / Table: ConstructConnect.

EXPLANATION: Table 3 conforms to the type-of-structure ordering adopted by many firms and organizations in the industry. Specifically, it breaks non-residential building into ICI work (i.e., industrial, commercial and institutional), since each has its own set of economic and demographic drivers.

Table 4 presents an alternative, perhaps more user-friendly and intuitive, type-of-structure ordering that matches how the data appears in ConstructConnect’s on-line product ‘Insight’.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect / Table: ConstructConnect.

*One in three Americans lives in one of the four shaded states, New York, Florida, Texas or California. Sum of first column may not exactly equal total due to rounding.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect /

Table: ConstructConnect.

* With respect to Tables 3 and 7, ‘transportation terminals’ is the one type-of-structure that is categorized differently in Canada (institutional) than in the U.S. (commercial), for reasons hav- ing to do with government statistics.

EXPLANATION: Table 7 conforms to the type-of-structure ordering adopted by many firms and organizations in the industry. Specifically, it breaks non-residential building into ICI work (i.e., industrial, commercial and institutional), since each has its own set of economic and demographic drivers.

Table 8 presents an alternative, perhaps more user-friendly and intuitive, type-of-structure ordering that matches how the data appears in ConstructConnect’s on-line product ‘Insight’.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect.

Table: ConstructConnect.

EXPLANATION: Table 7 conforms to the type-of-structure ordering adopted by many firms and organizations in the industry. Specifically, it breaks non-residential building into ICI work (i.e., industrial, commercial and institutional), since each has its own set of economic and demographic drivers.

Table 8 presents an alternative, perhaps more user-friendly and intuitive, type-of-structure ordering that matches how the data appears in ConstructConnect’s on-line product ‘Insight’.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect / Table: ConstructConnect.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect.

Table: ConstructConnect.

Source of actuals: ConstructConnect “Insight” / Forecasts: Oxford Economics and ConstructConnect.

Table: ConstructConnect.

Graph 19: Canadian Grand Total Construction Starts — ConstructConnect

Graph 20: Canadian Residential Construction Starts — ConstructConnect

Category: ConstructConnect

This entry was posted on Thursday, October 29th, 2020 at 5:02 pm. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.