Falling energy prices too much of a good thing?

Now for the bad news: Energy directly represents 8% of the S&P 500. Energy stocks also represent more than 20% of Canada's S&P/TSX Composite Index and probably an even greater portion of some overseas markets. And of course, banks lend a lot of money to the sector, too. For now, the asset price drops are mainly based on fear, but if low prices are sustained, there will clearly be more defaults than usual in the energy sector.

In addition, falling oil prices are yet another indicator of falling world economic activity. So while the U.S. consumer might be doing better, the world, in general, is slowing down. That will continue to reduce U.S. export growth. The relative strength of the U.S. economy and the higher interest rates that engenders will push the U.S. dollar ever higher. Consequently, U.S. goods could be even less competitive, and U.S. exports could slow even further. The dollar is now up well over 12% versus a basket of world currencies. Though tiny, oil-related jobs that had shown such growth this recovery are likely to at least plateau if not outright decline. Still, that is an industry measured in the tens of thousands of jobs, not millions. A recent report out of Oxford suggests that the U.S. economy had grown 9% since the recovery began, and shale oil- and gas-related projects accounted for 0.4% of that growth or less.

Retail sales burst higher in November

The retail sales for November were unusually strong, growing 0.7% (8.4% annualized) between October and November, and 5.1% year over year, indicating a strong start to the holiday season.

Increased savings, falling gasoline prices and better employment all provided the fuel for improved spending. The gains were broad-based, too, with only gasoline sales and the miscellaneous store category showing declines. Keep in mind that falling gasoline sales relate primarily to lower prices and not less demand. And lower gasoline sales mean that consumers have more cash left over to spend on other categories.

The report also showed a major revision to the October retail sales report, with the October growth estimate boosted from 0.3% to 0.5%. That makes for two great months back-to-back. This positive state of affairs suggests that the consumer sector will be a bigger contributor to GDP growth in the fourth quarter than it was in the third. A stronger consumer sector will offset what is expected to be a negative contribution from the net export sector.

Year-over-year data looks almost as good as the month-to-month retail sales numbers

Even the more conservative year-over-year averaged data shows a sharp acceleration.

On an inflation-adjusted basis, consumption growth has been rock solid at just about 2%. There was a slight dip in the first quarter of 2014 related to dire weather conditions in many key retail markets. Through the summer months, growth returned to the normal 2% level. Subsequently, growth has accelerated to 3.1% this fall. In fact, that is just one notch below the recovery high of 3.2% registered in late 2010 as the economy was just coming out of the recession. Note, looking at the non-inflation-adjusted line, the results don't look quite as positive, which is why a lot of people are being fooled into thinking that recent results are so lackluster. Indeed, low inflation is putting a lot of extra cash in people's pockets.

Retail sales are a very important component of total consumption, which is the No. 1 driver of U.S. economic activity, representing almost 70% of the U.S. GDP. Increased consumption generally means that businesses must boost production and employment to meet demand. Improved retail sales are the fodder for improved economic activity. Obviously, the larger services sector will play a role in that growth, but that services number is far less volatile and is driven by housing-related expenditures.

Retail sector data looks strong, too

Even the sector data was unusually strong with only the volatile gasoline sector and the catch-all miscellaneous retail category showing month-to-month declines.

There was at least a little more persistence in the sector results with strong sectors in October also showing strong results in November. That's a change from most of the rest of the year, when many categories turned up in the worst list one month only to turn up in the best list the following month. That made it a little had to call anything a trend.

Year-over-year data shows how much brick-and-mortar retailers are hurting

Year-over-year category data through October (the last month with the more detailed categories) shows that e-commerce sites, autos, restaurants and drugstores did the best. Department stores, jewellery stores, shoe stores and clothing stores all took it on the chin with declines or very small gains. A great deal of this is just rearranging the proverbial deck chairs with items normally sold by brick-and-mortar outlets moving to more online outlets.

Drugstores looking better, though not for economy-boosting reasons

Two standout categories are worth mentioning: restaurants and drugstores. These are both meaningfully sized categories that have been on a relative tear. The drugstore situation is a little complicated with shifting pharmacy management programs, (Walgreen (WAG) out and then in again with one major management firm) as well fewer drugs going generic in 2014. And the prices of some generics also moved up, aiding the drugstore number. Drugstore sales moved from year-over-year growth as low as a negative 2.4% in early 2013 to 5.4% currently and still trending upward.

Restaurants continue to stand out

Restaurants have been a huge story in 2014 after a horrific 2013. Year-over-year restaurant sales never got much above 1% from May 2013 all the way until December 2013. Now the food service growth levels are overreacting in 2014 in the other direction, with year-over-year restaurant sales up in the vicinity of 7%. And growth is uniform in the full-service and quick-service restaurants with both showing identical growth rates approaching the 10% level. Only lacklustre sales at drinking establishments are holding back the numbers from even better performance.

Better incomes justify the additional retail spending; good restaurant news should spread

That is a stunning performance and early indicator of sharply improving consumer confidence. Consumer incomes have been improving faster than spending for some time. A lot of that extra income is now finding itself back into the small splurge of eating out more. Quick-service numbers are also being helped along with better labor market reports. The correlation between fast food restaurants and labor market conditions is quite high. As consumers continue to gain confidence, as the labour market continues to pick up speed and gasoline prices remain low, I believe that consumer spending will remain robust for the rest of the holiday season. Bad weather last year means that without a repeat of last year's weather disaster in early 2015, retail sales comparisons will be like shooting fish in a barrel during the first quarter. (Bad weather pushed retail sales growth into negative territory last year.)

We stopped commenting on holiday sales reports for good reason

Careful readers may have noticed that I haven't been commenting on various short-term retail sales forecasts and even short-term results for this holiday season. This is neither an omission nor laziness on my part. I am sure today's national retail sales report for November surprised many, especially since at least one report suggested retail sales were down over 11% on Black Friday. However, anyone spending much time at the mall was not surprised by the healthy monthly retail sales report released this week.

The reason I started ignoring the various holiday metrics and forecasts is that a large number of firms found out they could get a lot of free publicity from reporting holiday forecasts or short-term results. This year consultants, software providers, retailers and even transportation companies weighed in on retail sales forecasts. Some days one could find one company projecting a horrible holiday season and another expecting a great holiday season on the very same page. Let's just say the methodology on some of these reports was less than robust. Exactly which days consumers would decide to shop and the role of online shopping have hopelessly muddled most short-term data. Getting things right on even a monthly basis with an army of government statisticians is tricky. The error rate in the monthly data from the U.S. Census Bureau is huge at the 15-day mark, when the data is often released. The second run at the 45-day mark is substantially closer to the mark. I have little faith in the instant analysis of a single day's data.

Small businesses building their enthusiasm

The National Federation of Independent Businesses has provided small-business sentiment data for several decades. Like a lot of sentiment surveys, it is at least a little biased by up and down movement in newspaper headlines as much as by real business conditions. For most of this recovery the survey has been a bit of a wet blanket, showing modest improvement, if any, most months. The index is finally showing some signs of life. The index jumped 2 full points for November, from 96 to 98. The index is now back to its long-term average. Gains in this multifaceted index were largely driven by improving business conditions and inflation-adjusted sales levels.

A lot of analysts ignore the headline data and most categories and head straight to the employment section of the report. The employment news for November was good but not great. The net hiring index moved up 1% to 11%, which is much improved from the recession but still below longer-term averages. The number of businesses saying that positions are hard to fill remained at 24%, one of its strongest readings of the past year. By itself the data doesn't mean a lot, but it does corroborate a lot of other positive labor market reports.

JOLTS holds its ground in October (written by Roland Czerniawski)

This week's Jobs Openings and Labor Turnover Survey showed that all three components--new openings, hires, and quits--were little changed in October. After large increases in new hires and quits in September, the most recent report suggests that no significant pullback has taken place, and that's good news. Currently, the numbers stand at 4.8 million, 5.1 million and 2.7 million for openings, hires and quits, respectively.

Given the large growth in November payrolls (up 321,000), it will be interesting to see the effects on the next JOLTS report. While it is difficult to tell with certainty, it could put a dent in openings and move the hires number upward. On the other hand, the November jobs report looked almost too good to be true. Eventual revisions to the monthly report may mean that the effects on the JOLTS data might be more muted than current data suggests. Also, it will be very interesting to see if the new hires came from the ranks of the unemployed and new entrants or from people quitting their existing jobs.

While openings and hires numbers are making decent improvements, quits recovery remains somewhat muted. We saw a relatively large month-to-month increase in quits in September, but outside of that one instance, this statistic has been treading water so far this year. Nonetheless, the number of quits is slowly approaching a prerecession range of roughly 2.8 million-3.0 million. As workers become more confident about the labor market conditions, the number of quits should continue to edge higher.

At the same time, this week's report showed that the number of unemployed workers per job opening continued to trend down, reaching a new recovery low of 1.86. A low number indicates a much tighter labor market with limited availability of workers. While this number might be more or less dramatic in certain industries, overall we are beginning to see a worker scarcity issue, which might become a much larger theme in 2015. In addition to the continued labor market improvements, some of the demographic trends, such as shrinkage of working-age population, might begin as early as next year. This would finally shift the wage dynamic in favor of workers. That is why the Fed is watching JOLTS so closely. Beyond just giving us a picture of the overall labor market conditions, it hints at the direction of the cost of labor, which in turn is a good indication of when substantial inflationary pressures may begin to arise.

Federal budget deficit continues to improve

October/November isn't necessarily a great time to look at budget data because neither month is a huge revenue or expenditure number. On the other hand, the December data includes some very important tax deadlines and is probably a little more representative. The good news is that after adjusting for timing of government payments due to weekends, the deficit managed to decease by $8 billion. That was true even in the face of a Medicare drug "true-up" payment in November. That was largely offset by $9 billion of fines collected from banks. Neither of those events is likely to recur. All of that aside, receipts grew 5.8%, while expenses grew a slower 2.3%. That caused the deficit, on a day-adjusted basis, to decline by about $8 billion year over year. That's a nice start to the year.

Recall that a year ago the Treasury received a large one-time payment of about $34 billion from Freddie Mac that is not likely to recur this year. Even considering the payment this year, the deficit should remain under 3% of GDP. The deficit should be a nonissue over the next five years, hovering around 3% of GDP, well below long-term averages. Higher interest rates and higher medical expenditures are expected to drive the deficit up to over 6% of GDP by 2039, so the United States is not out of the woods just yet. However, that deterioration assumes much higher inflation and higher interest rates that just may not happen.

The consumer and business balance sheets remain in great shape

While not a particularly useful forecasting tool, the Federal Reserve's Z1 report provides a comprehensive look at business and consumer balance sheets. I always like to look at overall consumer net worth, which takes all assets and subtracts out debt. In 2007, before the recession hit, assets totaled about $66 trillion before falling to $56 trillion in 2009. Since 2009, assets have increased $25 trillion to $81 trillion as of September. To put that in some perspective, currently the GDP level stands at $18 trillion. So assets are many times larger than GDP and the recovery gain in net worth exceeded the annual run rate of GDP. (I repeat, run rate--not growth rate.) That has certainly provided a little fuel for the recovery. Though, only a small portion, 3%-5% of asset gains, is typically ever spent. Unfortunately, those asset gains are beginning to slow a bit as the stock market gains have slowed, as has the rate of home price appreciation--the two big drivers of the net worth calculation.

The report also showed that consumer debt growth has remained exceptionally muted. That has clearly held back this economic recovery, perhaps more severely than many more widely quoted pieces of data, including hourly earnings. From the 1980s to the early 2000s, very modest wage growth--not much better than levels seen today--was supercharged by loan growth and parlayed into economic activity that looked quite a bit stronger than it is today. At the epicenter of debt creation, consumer debt grew by more than 10% each year between 2004 and 2006 before slowing to 7% in 2007. Debt growth then turned negative for three years and remained at just 1.6% in 2012 and 2013. That is less than the rate of inflation. Consumers look a little optimistic in 2014, with loan growth of 2.7% on an annualized basis. It is no accident that economic activity has picked up at least a little lately as some additional debt is supplementing wage growth.

Industrial production and housing data due next week along with Fed meeting

Industrial production is expected to look quite good in November based on improved hiring data in general and cold weather that is likely to drive up the volatile utility sector. The index is expected to be up 0.9% versus a 0.1% decline last month. Sector-by-sector analysis will continue to be important. I am expecting a bounce in auto production.

Housing starts are expected to be little changed at 1.03 million units. Not much has changed recently to suggest improvement or deterioration. Better employment data suggests better results next year.

The Consumer Price Index will also be released next week with expectations that prices in November will show an outright decline of 0.1% due to falling gasoline prices. I will be watching other categories, especially drugs and food, for any signs of acceleration.