Worldwide market activity this week was more related to geopolitical events than any major economic news. The alleged shooting down of a Malaysian jetliner and more Palestinian rockets fired into Israel followed by Israeli incursions into Gaza were all unsettling, to say the least. The market reacted as one might suspect on Thursday, falling sharply.
Still, that didn't last long with markets on Friday gaining back a hefty part of what they lost. U.S. bonds rallied nicely this week as they got a lift from those seeking a safe haven in a very unsettled world. News that China has been buying a lot of U.S. Treasury bonds in 2014 didn't hurt bonds, either. Apparently, unusually strong bond buying by China has been one of the previously unknown reasons U.S. bond yields have dropped so far this year. By buying those bonds, China effectively forced the yuan lower, which helps China's exports to become more competitive. It also reduces U.S. interest rates, which stimulates U.S. economic activity and usually drives more Chinese imports.
Apparently, some of that move (along with changes in lending rules and active government encouragement for banks to lend more money) are beginning to work. China's economy grew 7.5% year over year in the second quarter, which is better than expected and arrested a recent decline in economic activity. Although demographics are likely to drive growth to 5% or less over the next several years, the dike has been plugged for now, dispelling worries of a hard short-term collapse in the Chinese economy. News of better economic growth in China moved emerging markets higher this week despite all the geopolitical turmoil. In other world economic data, the eurozone's year-over-year inflation rate for June was 0.5%, the same as it was in May. Markets felt greatly relieved that at least inflation stopped going down, and excluding food and energy, inflation was slightly higher.
U.S. economic news was almost uniformly bad when comparing May and June numbers but continued to improve on a year-over-year basis. Retail sales looked sloppy at just 0.2% growth between May and June, but that was strongly affected by volatile auto and building material stores. Year over year, the retail sales data showed its fourth straight month of improvement. Likewise, industrial production numbers for manufacturing look soft with just 0.1% growth, but year-over-year averaged data showed its fourth month of improvement and was well above long-term averages. Housing data remained in the doldrums, but better homebuilder sentiment, falling mortgage rates, improved weather and much better employment data should mean better days ahead.
Corporations seem to be losing interest in investing for the future
Lately I have been complaining about the skills mismatch between workers and employers. This article from The Wall Street Journal highlights some of those issues and what one firm is doing about it. Just as important, the article notes the recent neglect of employee training and development programs. Unfortunately, it's not just employees that corporations have been neglecting, but investing in general. There seems to be a greater interest in merging with competitors and better yet, an overseas transaction that has the potential to reduce taxes.
Poor month-to-month retail sales report overshadowed by continued year-over-year gains
Recently, interpreting the retail sales report has become a real challenge. Terrible winter weather, the subsequent huge bounceback and abnormally large data revisions have complicated an already difficult task. Some recent revisions, just one month later, have been much larger than the originally reported data point for overall retail sales growth.
Consumer doing just fine, thank you, with more good news ahead
Before diving into the details, I want to be very clear: Retail sales are improving and have been doing so for a while after a terrible winter. Increasing employment (June data was the best of 2014) and income data suggest that retail sales could accelerate even more from here. Weekly shopping centre data for reporting periods after the June monthly data was released suggest even more gains. Single-point year-over-year shopping centre sales are up 4.5% for the week ended July 12 and 3.9% on a five-week averaged basis, according to the International Council of Shopping Centers. Fading high winter utility bills and some relief from the droughts that spiked food prices should begin to provide some further fuel for consumer spending in the months ahead.
On the surface, retail sales gains of just 0.2% between May and June were a disappointment compared with hopes of 0.6% growth. However, April and May sales were revised upward by a combined 0.3%, which means that the raw dollars of retail sales for June were just about as advertised. Also, excluding autos and gasoline (two highly volatile sectors of this report that are not used to calculate GDP; those data points come from a different report) retail sales were up a more respectable 0.4% (4.8% annualized).
By sector, the news was quite good with a lot of key sectors doing well and only a few doing poorly. Performing above the overall average month to month were general merchandise, drugstores, apparel, electronic retailers and grocery stores. These are bread-and-butter categories that I am glad are doing so well.
On the loser side, building materials looked seemingly awful, falling a large 1.1%. However, the context is that this category has been up 2.3%, 1.4% and 0.6% over each of the previous three months, probably reflecting a rebound from difficult winter weather conditions. A relatively flat housing market and a wet early summer season in the East and Midwest probably didn't help matters, either. The auto sales decline remains a bit puzzling as overall unit sales were up meaningfully in June. Mix issues (business and fleet auto sales versus personal purchases), price issues and timing of revenue issues all complicate the comparisons between reported retail sales of autos in this report and the factory-reported auto sales we saw several weeks ago. Fortunately, the more favourable factory numbers are used in the GDP calculation and auto sales should be a nice positive to the GDP report. Food service and drinking places (restaurant sales) were also off, but this category has also been highly volatile lately with sales up a strong 0.9% in May before slumping 0.3% this month.
Year-over-year data averaged paints a much better picture of the retail environment
Year-over-year data averaged shows sales up by 3.9% versus a 12-month average of 3.4%. Better yet, that metric has improved four months in a row and is now at its highest level of the past 12 months. Recently improved retail data suggests that consumption in the June quarter should be higher than it was in the March quarter, potentially providing a small boost to the second-quarter GDP report.
Housing starts and permits data remain baffling
The headline number for housing starts in the U.S. was disappointing, with a 9.3% month-to-month decline in starts with the pain just about equally spread between the more important single-family segment and multifamily homes. The silver lining is that June starts at 893,000 were up substantially from last year's poor showing of just 831,000 units. Much like the retail sales data, the year-over-year housing starts report is much better and has been improving for several months, as shown below.
There was another anomaly that is difficult to explain in the starts report. The Southern region reported a more-than 30% decline in starts while the other three territories were all up. The Midwest, Northeast and West were up 28%, 14% and 10%, respectively. It's unusual to see one region of the country moving this contrarily to the rest of the United States. To me, it seems like a data collection error that might eventually be revised away. Other news outlets, including The Wall Street Journal, are saying the rainy spring was the culprit in poor Southern district starts, as the weather limited both site preparation and new framing activity. Living in the Midwest, I can tell you things were even rainier here, but the Journal correctly notes that land is effectively stockpiled in the Midwest during the cold winter months. This provided Midwestern builders a built-in cushion that year-'round builders in the South just don't have.
All of this said, year-to-date data suggests that the housing economy has not made as much progress as hoped. The litany of excuses is huge, including higher mortgage rates, higher prices, lack of land, slow government approvals, tight lending conditions and large labour shortages.
But looking forward, mortgage rates are down from the beginning of the year, the weather has improved dramatically, and the employment market has been much improved. Even lending conditions are a little better. Better yet, the single-family market that generates substantially more economic activity is acting better at last.
Homebuilder sentiment signals better days ahead
Homebuilders seem to be sharing some of my optimism as a recent builder sentiment survey showed its best reading in five months at 53. The reading means that more builders are seeing improvement in their business than those seeing declines.
This is the first time that I have been able to say that since January. All three parts of the report, current sales, future expectations, and traffic were improved over May's levels with future expectations showing the most improvement and traffic showing the least improvement (and still at a disappointing level of 39).
Interestingly, by region the South was at 51, the second best of the four regions, which suggests that something wasn't quite right with dismal report for Southern housing starts noted above.
Industrial production ekes out another monthly increase, year-over-year data continues to impress
Overall industrial production that includes manufacturing, utilities and mining was up 0.2% between May and June. The manufacturing-only segment, which is the most important sector, was up a relatively slow 0.1%. Perhaps the manufacturing sector was one of the most affected by bad weather. A one-day plant shutdown due to weather basically means 5% of monthly production is lost (although overtime on remaining days can make up some of the difference). A massive decline in January was followed by two stunningly good months with performance since then being more normal. Manufacturing industrial production on a year-over-year, averaged basis has been up for four months in a row. At 3.7% currently, the index is well above its average since 1972 of 2.6%.
Sector data didn't add a lot of insight to our analysis, although entities serving the auto industry (but not the auto manufacturers, which were down for some reason) were generally the strongest performers. Aerospace also notched a second great month, growing 1.1%. The big detractors were petroleum and coal products (negative 2.7%), apparel (negative 1.3%--too much inventory), food (negative 0.6%--consumers buying less because of high prices) and motor vehicles (negative 0.3%--big boosts in prior months). Overall, looking at the category data I would have guessed the report would have been better, but a couple of bad apples really spoiled the short-term story.
More housing news, consumer prices and durable goods orders due next week
Hopefully, next week's housing data will look better than this week's. Both existing-home sales and new home sales will be released next week. An improving pending home sales report for May should point to a nice uptick in June and July existing-home sales. The consensus is for existing-home sales to finally pop back up over 5.0 million units in June (compared with 4.89 million annualized units in May) after dipping as low as 4.59 million units in March. That will still put it modestly below last year's 5.1 million units, but that was strongly aided by rising mortgage rate fears. I am also hopeful that inventories begin to pick up some, which has been a major constraint in some markets.
New home sales will likely drop from a too-good-to-be-true 504,000 annualized units in May to 475,000 units in June. While single-family housing starts have improved very modestly lately (according to this week's report) they weren't nearly as strong as last month's separately reported and calculated new home sales report. With new home sales averaging just 431,000 units over the past 12 months it is hard to argue that new home sales (which included started, completed and still-on-the-drawing-board homes) will remain over 500,000 homes. In fact, a downward revision for May would not surprise me.
Consumer prices are expected to have another not so great month with prices expected to be up 0.3%-0.4% (3.6%-4.8% annualized). Year-over-year averaged prices should still remain very near 2%. Seasonally adjusted gasoline prices will hurt the report, but I am thinking that some food price increases may begin to moderate in the June report. With drought conditions improving across much of the United States, I suspect food prices may begin to decline by the end of the year. Auto prices may begin to moderate, too.
Durable goods numbers have been quite volatile lately. Last month overall orders fell 0.9% and this month the consensus estimate is for no up or down movement. Excluding airliners, both May and June orders would have been higher. The more important year-over-year growth in nondefense capital goods orders has been improving for several months (a key measure of business confidence).
Analysts will also be carefully watching Markit purchasing manager flash results. A lot of China data this week showed signs of bottoming and a good flash PMI number for July would affirm that improvement. Also, the European PMI data weakened some last month. That was followed by some very poor industrial production numbers for May that greatly worried investors who fear another European slowdown. Hopefully, the European PMI data will look a little better when released next week.
