Weekly Market Update - August 4, 2016

Construction Put-in-Place (Census Bureau): Total construction (NSA), in June recorded growth of 12.1% y/y to $1156.2 Bn with $829.8 Bn from the private sector, up 16.2% y/y and $305.5 Bn from state & local, up 3.6% y/y, (Table 1). On a three month year on year basis, growth was slower with total construction up 4.6%, private 9.1% and state & local down 3.8%. The result was negative momentum of -7.5% overall, -7.1% for private and -7.5% for state & local construction. Looking down the list we see negative momentum for all but two project categories; Power and Conservation.

Non-residential constructional (NSA) recorded strong numbers in June, up 12.8% y/y overall, with Private NRC surging 16.0% y/y to $393.9 Bn and State & Local up 4.9% y/y to $141.9 Bn. On a 3 month y/y comparison, growth was slower with overall, up 6.2%, private sector up 9.6% and state & local contracting 2.5%, (Table 2). Momentum therefore was negative across the board, indicating that the growth rate for NRC is slowing somewhat. The downward bias matches with the Dodge starts numbers (last week’s Gerdau market update), which have been trending lower for eight of the past 12 months y/y.

Figure 1 presents a chart on non-residential building construction expenses from 2010 to present. The good news is that we continue to see y/y growth, the bad news is that the rate of growth is rapidly decelerating.

Total infrastructure (measured as a rolling 12 month total), includes, Highways, Streets & Bridges, Transportation, Sewage & Waste, Water Supply, and Conservation has declined in each of the last three months in expenditures terms. In percentage terms total infrastructure has increased for 50 consecutive months, but as Figure 2 illustrates the rate of growth is falling, down three months in a row.

Table 1

AIA Consensus Non-residential Construction Forecast: Twice per year in January and again in July, the American Institute of Architects canvases seven leading forecasting firms looking for each company’s prediction for non-residential construction (NRC), and its sub-categories for the next two years. The forecasting firms include: Dodge Analytics, IHS-Economics, Moody’s, FMI, CMD, Associated Builders and Contractors and Wells Fargo Securities LLC. Table 3 lists these predictions. Yellow highlighted numbers represent the highest estimate for the year/category, and blue highlighted numbers represent the lowest estimates. Dodge Analytics offers by far the most optimistic forecast, while the Associated Builders and Contractors has the most pessimistic forecast.

The AIA consensus forecast, estimates 5.8% y/y growth for all NRC 2016, followed by a 5.6% increase in 2017. In its January forecast, these numbers were 8.3% for 2016 (down 1.1 points  in July over January) and 6.7% for 2017, (down 2.5 points in July over January).

Breaking the current forecast down into its major sub-categories, the July consensus forecast for Commercial construction is 11.7% (higher than the 9.9% January estimate) in 2016, falling to 6.5% in 2017 (much lower than the 10.4% January estimate). The consensus forecast for Industrial construction is -2.1% for 2016 (down sharply from the 11.9% estimate in January) and 2.9% in 2017, (down from the 5.3% January outlook). The Institutional construction consensus January forecast was 6.7% for each year. Both were revised downward to 4.5% in 2016 and 5.8% in 2017.

The range of estimates illustrates just how perilous the “science” of forecasting is. There are some extreme variations in forecasts from firm to firm, as depicted by the “yellow and blue” highlighter on each line. In the most extreme case in 2016 with half of the year “in the books”; Moody’s predicts 22.7% growth for Amusement/Recreation while IHS forecasts negative 2.2% growth for the same project category. For 2017, the same two forecasters’ predictions for Religious buildings range from negative 10.1% (IHS) to positive 14.2% (Moody’s).

Table 3

Gross Domestic Product 2nd Q 2016 1st Estimate: On July 29th the Bureau of Economic Analysis released the 1st estimate of GDP growth for the 2nd quarter of 2016 with revisions back to Q1 2013. The reported growth rate of 1.22% failed to meet analysts’ expectations. GDP is measured and reported in chained 2009 dollars and in the 2nd estimate of Q2 was $16.575 trillion. On a trailing 12 month basis, GDP was 1.23% higher in the 2nd quarter of 2016 than it was in Q2 2015. In the last six years, since Q1 2011 the trailing 12 month growth of GDP has tracked between a high of 2.88% to a low of 1.07% therefore the latest result remains within this range at the low side.

Figure 3 shows the headline quarterly results since 1990 and the latest IMF forecast through 2020. In their April revision, the IMF downgraded their forecast of US growth in 2016 from their October estimate of 2.84% to 2.40% and downgraded 2017 from 2.80% to 2.50%. It looks as though the IMF may still be optimistic.

Figure 4 shows the change in the major subcomponents of GDP in Q2 2016 and the dominance of personal consumption as a growth driver. Personal consumption accounted for 69.3% of total economic activity in Q2 and contributed 2.83% to GDP growth which was the highest since Q4 2014. Personal consumption includes goods and services, the goods portion of which includes both durable and non-durables. Retail sales again surprised on the upside in June, suggesting consumer spending growth may be gaining momentum. The change in inventories was the greatest detractor in Q2 with a contribution of negative 1.16%. Declining inventories have a negative effect on the overall GDP calculation and this has been the case for the last five quarters. Over the long run inventory changes are a wash and simply move growth from one period to another. In this latest data the negative contribution of inventories was the greatest since Q1 2014. Construction, both residential and non-residential made negative contributions in Q2, which is at odds with the Commerce Department report of construction expenditures. The contribution of fixed residential investment at negative 0.24% was the first time this component detracted from GDP since Q1 2014. The contribution of fixed nonresidential investment which includes infrastructure has been negative for three consecutive quarters. Net exports made a positive contribution in Q2 and added 0.23%. Government expenditures contributed negative 0.16% in Q2 down from positive 0.28% in Q1. Figure 5 shows the contributors to GDP extended back through Q1 2007 and describes the quarterly change in the six major subcomponents. Figure 6 shows the breakdown of the $16.6 trillion economy.

Figure 3

Regional Job Creation through June 2016: The regions have fared very differently since the pre-recession high of 1st Q 2008 and since the low point of 4th Q 2009. There are now 6,471,000 more people employed than there were immediately before the recession, but of that number, almost half, occurred in the South Central, (TX, OK, LA and AR) and Pacific regions, (CA, OR and WA), (Table 4). The 3rd Q of 2015 was the first time that every region had more people employed than it did at the pre-recession peak. Employment is now 14,901,000 positions higher than it was at the low point of the recession. The Pacific has had the largest number of jobs created during the recovery, with an increase of 3,017,000 positions which amounts to 15.4% of total employment. The South Central is in 2nd place with 2,093,000 jobs created which is 14.0% of total employment. Table 5 shows the number of people employed in each region. The East North Central, (IL, IN, MI, OH and WI), currently has the highest number of employed people with 21,833,600 but the Pacific with 21,540,000 is catching up fast. During the depths of the recession in Q1 and Q2 2009, the East North Central had the highest number of job losses

Table 6 shows the history of quarterly job creation by region for the last 5 years. All regions except the East South Central, (AL, KY, MS and TN), had positive job creation in Q2 2016. The effect of energy prices is evident in these regional employment reports. The South Central region went off a cliff in Q1 2015 after having the highest job creation of any region in the previous quarter. This region rebounded in Q2 2015 through Q4 2015 but has since declined again and is nowhere near where it was before the oil price collapse. In the six quarters since Q4 2014, North Dakota lost 22,800 jobs, Oklahoma has lost 4,300 jobs but Texas gained 213,300 jobs. The Pacific region created the most jobs in Q2 of this year followed by the South East, (FL and GA). The East South Central lost 5,300 jobs in Q2 2016 mainly in Mississippi which was down by 10,000. Alabama was up by 7,300.

Figure 1

U.S. Housing: Total housing starts were reported at a seasonally adjusted annual rate of 1,189,000 for the month of June, an increase of 4.8% from May, but down 2% from June 2015. Tracking the three month moving average (3MMA) to smooth out the month to month volatility in the data, the June 3MMA was 1,160,000 units, (Table 7). Single family housing starts rose 7.6% 3MMA y/y, while multi-family fell 11.2% 3MMA y/y. Housing in the northeast and the western US were weak, while ground breaking for both single family and multi-family boomed in the mid-west and southern US, (Figure 7).

A forward looking gage of future housing construction can be gained by examining housing permits data. On a 3MMA June housing permits fell 9.5% 3MMA y/y, with permits for multi-family construction falling 29% 3MMA y/y. Permits for multi-family housing in the northeast and southern regions were anemic, with single-family housing permits continuing to rise in all regions of the US. The mid-west had the strongest demand for residential housing with permits for construction and starts both showing growth. Foreclosures have fallen for 3 consecutive months, (Figure 8) down 6.3% m/m in June, as the Months of Supply for new and existing homes also tapered off, (Figure 9). Data from the National Realtors’ Association showed there are 8.7 existing homes for every new home sold, with the average price for a new home plus land at $358,200, and the average price for an existing home plus land at $292,100, (Figure 10).

Table 7

Preliminary U.S. Steel Long Product Imports (SIMA): June preliminary imports of steel long products rose 13% from May final imports, an increase of 65,000 tons (kt) month / month. Year to date the US has imported 3.06 million tons of long products, down 7% compared to total imports for the six months through June 2015, (Table 8). Tracking of import licenses is a forward looking data set that gives insight into expected imports each month. Based on US import administration rules, countries applying for permission to import have up to 75 days to use licenses that were applied for, or lose those licenses. As well, global exporters have up to the 10th of the following month to apply for licenses, e.g., exporters had until July 10th to formally apply for June imports.

Preliminary rebar imports rose 39% m/m as 17 kt of requested licenses were not fulfilled during the month of June. Japan and Spain were recorded importing 6,000 to 7,000 tons more than originally requested for June. In contrast, Turkey (17 kt) and Kazakhstan (11 kt) applied for thousands of tons of rebar licenses that were not utilized, and could be imported during July or August. Year to date, rebar imports were flat, (down 1% y/y). Preliminary wire rod imports fell 13% m/m. In spite of this month over month decline South Korea also imported more wire rod than originally requested, to the tune of 5,200 additional tons.

Structural steel imports rose 19% m/m, with preliminary tonnage totaling 99 kt. The steel beam component of this product category remaining flat m/m, representing increased tonnage of heavy angles and channels. Year to date heavy structural steel imports, plus June preliminary totaled 499 kt, a decline of 5% y/y. Light shapes preliminary imports, as defined by the Steel Import Monitor, fell 17% m/m, yet remained flat compared to imports through the first six months of 2015, down 2% y/y.

Preliminary imports of hot-rolled bars (HRB) and cold finished (CF) steel bars combined encompass Gerdau’s definition of special bar quality (SBQ) and merchant bar quality (MBQ) steel products. By reviewing the past several years of final import data compared to requested licenses, we made an educated calculation of probable MBQ & SBQ imports. Through 2013 and 2014, on average, 10%-15% of HRB were MBQ products with the remainder classified as SBQ. Since mid-2014 this has risen and now averages 20%-25%, (Figure 11). Using this extrapolation, we expect MBQ imports of approximately 22 kt for the month of June, a 12% m/m increase compared to May final imports. Year to date, MBQ imports, plus preliminary calculations, were down 11% y/y. SBQ preliminary imports rose 13% in June, yet are down 30% compared to imports through the first six months of 2015.

Table 8

Figure 11

Market Trends for Steel Long Products: Twice each month in this report we publish our findings on the present situation of the long products market and where it is headed. We also keep track of how these two data sets are changing over time and we compare to the situation in August 2008, just before our businesses went off the recessionary cliff.

Figure 12 shows the trend of our present situation analysis monthly since January 2010. Our view of the present situation progressively deteriorated from January last year through February this year by which time we regarded only 7 of 26 indicators as being historically good. This deterioration did not translate into a comparable increase in negatives, rather the number of indicators that we regarded as historically normal took up most of the change. In other words things were just ho-hum and not terrible. Since December of last year our view of the present situation hasn’t changed much. The more important aspect of this analysis for corporate management is where the situation is headed. Figure 13 shows the trend of the trends also since January 2010. There is nothing subjective about the trend numbers, these are facts released in each month and the change in those values. The number of indicators trending positive peaked at 85.2% in August of 2014 with 11.1% trending negative and one indicator unchanged. There was a gradual deterioration through January this year by which time negative trends exceeded positives in the ratio 46.2% to 53.8%. This was the worst result in over 3 ½ years. February this year saw a small improvement but in March the number of indicators trending positive surged to 65.4% and held that proportion through May. There was a small decline in positives in June and July when positives were 60%. We regard this data analysis as relatively stable therefore it looks likely that a positive change is really happening. It’s worth noting that at no time since January 2010 have we ever come close to the trend situation that existed before the crash in 2008. We really should have seen that coming but few did.

Figure 12

Contributors this week include; Laura Remington, Peter Wright and Steve Murphy