The market's absolute fixation on the U.S. Federal Reserve continued this week. Stocks, bonds, and even commodities all soared this week as it appeared the Fed was adopting a more dovish tone at its latest meeting. European and emerging markets did almost twice as well as U.S. markets with weekly gains approaching 5% outside the United States. The Fed pulled in both its economic forecasts and individual interest rate projections. This seemed to reduce the odds of a Fed rate increase at its June meeting, with most forecasters now believing rates will be moved up in September, and some even suggesting a rate increase may be off until 2016. Again, all of this would just be delaying the inevitable and setting us up for an even bigger fall once it does begin moving rates up.
On the economic front, the market didn't seem too upset that both housing starts and industrial production, two key components to the recovery, didn't look so hot at first blush. On the good news front, Greece seemed to take a more conciliatory stance toward the EU, which seemed to assuage some market fear this week and is perhaps one of the reasons European markets finally crushed performance in the United States.
The Fed blinks
Every time it seems as if the Fed is moving toward a rate increase, the rug seems to be pulled out from underneath it. The Fed statement after its March meeting suggests that it is worried about the pace of the economy and the impact of the strong dollar. Inflation doesn't appear to be much of a worry to it. Unfortunately, the market had already built in and assumed that strong economic conditions are likely to force the Fed to raise rates as early as June. Now everyone is unwinding those positions. Now markets are likely to go into another funk as economic data begins to improve and the next Fed meeting approaches. I am growing a little weary of this on-again, off-again pattern and will spend my time focusing on underlying economic fundamentals. That's what's driving the Fed anyway.
The Fed moved its economic forecast back from a range of 2.5% to 3% to a range of 2.3% to 2.7% for 2015. That was a forecast that we always found to be too high. Our own forecast is still even lower at 2% to 2.5%, with growth in the first quarter potentially at the low end of that range. Oddly, the Fed dropped its unemployment forecast to 5%-5.2% from 5.2%-5.3% at the same time it lowered its economic forecast. That doesn't make a lot of sense to us.
We agree that a lot of short-term month-to-month data, including retail sales, industrial production, and housing starts, do look suspiciously soft. The year-over-year data suggests a plateauing in growth, not a decline, and that there isn't a lot of reason to worry. Weather conditions have certainly managed to mess up economic reports yet again, artificially depressing results in the short term. As the bad weather lifts, I suspect the economy will accelerate again, paradoxically causing the market to panic once more over the timing of the first rate increase.
In my mind, the Fed can't play with its rate decision much longer if it wants to take a slow and gradual approach to rate increases. While inflation remains low now, the labour shortages we are forecasting and a rapidly closing output gap could make inflation an issue again sometime between late 2016 and 2018. Waiting until the inflation is fully visible would be a huge mistake, in our opinion.
One trend a bit worrisome to us and the Fed is that employment growth seems to be accelerating at the very time economic growth is taking a pause. Sometime in the near future, economic growth will have to accelerate or employment growth will need to slow. Businesses don't hire workers for the fun of it. Annual budget and hiring cycles that are hard to back away from, especially early in the year, suggest that it may be the hiring portion of the equation that needs to change.
A closer look at housing data suggests some improvement
Unfortunately, media outlets are focusing on the 17% month-to-month drop in housing starts for the month of February in isolation. The gloom-and-doomers are now piling on, contending that by combining this week's industrial report, last week's retail sales report, and now the housing report, one might guess the U.S. economy is losing serious steam. In previous columns we warned not to expect much from these reports, which volatility and weather are battering. A lot of outlets are at least mentioning the weather issue. And the effects are real. On some level, the worse the weather in the short run, the bigger the bounce will be this spring and summer.
Less mentioned is the fact that month-to-month numbers have been highly volatile instead of trending in one direction or the other. Lately, many statistics have surged for a month or two and then shown little net monthly change, or even some declines. However, looking at year-over-year averaged data, it's possible to see some stability and maybe even some improvement.
Looking specifically at housing again, starts were indeed down a seemingly meaningful 17% month to month. However, year-over-year, averaged data, which includes bad Februaries in both years, seems to be showing some improvement in both starts and permits.
Even though the year-over-year starts data look similar to permits (7.2% versus 6.7%), the permits data is more impressive for its consistency. I have taken a real shine to permits data because it tends to be less volatile than starts data and is better at staying on the right side of trends. We note that a permit on a home can be pulled, even in relatively bad weather. However, if the ground is too frozen, snow-covered, or muddy, the actual start may need to be delayed. Note that permits never turned negative in 2014 even with record-setting weather conditions, but housing starts did. Starts growth rates got into the double-digit range over the summer; permits never made it out of single digits.
Permits data well above starts, growing, and at a new recovery high
In units, permits hit 1.09 million units, the best number of this housing recovery and up about 3% month to month. That's not a bad report for a month with admittedly terrible weather. Because permits were so far above starts, 1.09 million versus 0.89 million, starts would seemingly have just one way to go: back up.
Weather certainly limited starts
The relationship between weather and the starts data showed up in spades in the regional starts data, too. Starts in the hard-hit Northeast territory (which includes Boston) were down over 57%. Starts in the Midwest were down 37%. Things were better in the West, which shows a smaller 18.2% decline, while the South, by far the biggest housing market (more than 50% of the total) was down by just 2.5% in February compared with January.
Both starts and permits, in units, were above their 12-month averages in January. February's starts plunge puts it below its average, but less volatile permits are still in good shape. Still, we aren’t lighting the world on fire just yet as permits in both categories have been relatively stagnant over the past several months. That doesn't trouble me much. It's another case of data that jumps one month and then just sits. Longer term, acceptable affordability, easier credit, and millennials who are finally beginning to make purchases that were delayed because of the recession should move both starts and permits upward.
Homebuilders still optimistic (by Roland Czerniawski)
While the homebuilder sentiment ticked down 2 points in March, the sales expectation component of the survey remained relatively high at 59. It is the buyer traffic that remains depressed at 37, but some homebuilders are beginning to see improved conditions. 
Lennar LEN, a leading U.S. builder, reported its earnings results this week. The Wall Street Journal has done a nice visual summary of those results; the gist of it is that the results looked encouraging and that residential construction could show signs of life going into this spring season.
More important, we are now seeing more confirmation of the hypotheses we've been writing about over the past few months. First of all, it's great to see that the price for new single-family homes is finally coming down, which is something we thought had to happen in order to better suit the growing cohort of younger buyers. That also should narrow the huge gap between pricing for new and existing homes. Second, as we suspected, the slump in oil prices has unfortunately begun to weigh on the Texas region and its large housing market. Lennar reported that its sales in Houston are down 10% and 15% year over year in January and February, respectively. Finally, the strong preference for renting seems to continue, and it has typically been a detractor from the single-family residential construction. Lennar, however, announced that it will experiment with building single-family homes for rent. If other builders were to follow such a strategy and introduce it on a large scale, it could provide a meaningful boost to the residential construction sector. For now, the impact of this initiative will probably be minuscule, and it signals continued strength in the rental market.
Demographics mean more first-time homebuyers
Demographics, at least relative to first-time buyers, are improving, too. First-time homebuyers are typically around 31 years old. That age cohort shrank from its peak of over 4.3 million people (those born around 1960) to a low of 3.2 million (those born in 1976). It's no coincidence that the housing market was in a funk in 2007 when that small-size age cohort from 1976 was turning 31. The number of native borns turning 31 in 2015 is now back up to 3.7 million. In very round numbers, we are halfway back to the birth peak. Although births for a single year aren't likely to get back to their baby boomer highs, they should get very close. New 31-year-olds should steadily increase each year through 2021, recovering to 4.2 million potential first-time buyers. That group represents the echo of the baby boomers. Over the past five years, that nicely expanding age cohort was in its 20s, helping explain some of the interest in apartment living and strong multifamily starts and permits.
Industrial production disappoints; again, no need for despair
Industrial production grew a paltry 0.1% overall between January and February, falling below expectations for growth of 0.3%. Just as disappointing was a revision to the January data that now shows a 0.3% decline instead of a 0.2% increase.
These disappointments come despite the fact that utility demand surged by over 7% in a single month because of cold weather. Excluding utilities, which make up 10% of the index, industrial production would have been down 0.6%.
The utility section is one that I generally like to ignore, because it doesn't tell us anything about the strength of the economy, just how cold it was. And generally what goes up must come down, so next month utilities could be a huge drag on the index. Plus, utility demand doesn't generally do much for utility-based employment, either. Also, if consumers are spending more on utilities, they won't have the cash to spend on other goods and services. A lot of people are asking why the consumer isn't doing better with gas prices being so low, and part of the answer might be high utility bills. Unfortunately, the lingering effects of high utility bills can last two or three months given long billing cycles. February natural gas may not need to be paid for until April or even later.
Offsetting the strong utility data was the mining sector, which slumped 2.5% and accounts for almost 18% of the production index. Some pundits are attributing the poor showing to oil and gas production slippage. That would be wrong. Oil and gas extraction in February was up 0.3% from January. We suspect that number will eventually begin to fall, but not yet. The real shortfall came in coal mining, which was down 7%, and scattered declines in other non-energy-related categories. The supports services for both regular mining and oil and gas production were both very weak, too. That will eventually turn up in the production numbers.
The manufacturing sector, which is the most important part of the report, was also soft, declining by 0.2%. This marked the third monthly decline in a row for this important metric. However, I caution that this series is prone to big bumps and plateaus instead of nice, steady straight-line growth that confounds a lot of econometric models. The year-over-year data has been considerably less volatile. It shows a flattening growth pattern that is likely to turn into a declining growth rate pattern in the months ahead. For the full year, industrial production is likely to slow to 3%-4% in 2015 from 4.4% for all of 2014. Continued slowness in the housing industry, slowing growth rates for exports and autos sales, as well as a slower oil and gas business will all hold back manufacturing in 2015.
Month-to-month sector data shows broad-based weakness
The weakness in manufacturing output was broadly based with more than half of all industries showing a decline between January and February. Aerospace, computers, textiles and petroleum products were the only categories to show gains. Autos, primary metals, and apparel were all particularly soft for the month. While helping utilities, poor weather likely played a role in the broad-based declines. Supply disruptions related to issues at the West Coast ports probably didn't help matters, either.
On tap next week: homes data, CPI, manufacturing data and final GDP reading
Existing-home sales are expected to show a little improvement from 4.8 million units in January to 4.9 million in February. These are still well below recovery highs of around 5.4 million units and are likely being affected by harsh weather conditions. Given some of the other housing data we have seen, the 4.9 million units might be a bit of a stretch. However, the worse the February numbers are, the better the numbers are likely to be later this spring.
We are finally expecting a little inflation in February. After a couple of months of deflation in gasoline prices, the primary driver of the deflation managed a healthy increase in February due to refinery issues and temporarily higher crude oil prices.
Durable goods orders and Markit purchasing manager data could give us a little hint about the health of the manufacturing sector. These numbers have been flat to trending down lately. I am a little afraid that a slowing energy sector is likely to keep a lid on both of these reports. The expectation for the durable goods order report is a gain of 0.6%, but that does include those volatile airline orders. Poor weather and port strike issues may be factors in February. Those factors are likely to reverse themselves in March and April as ports get back to normal and the weather breaks.
The final report on GDP for the fourth quarter is also due next week. Due to higher health- and insurance-related issues, seen in a separate U.S. Census Bureau report, those figures will need to be raised in the GDP report. However, it is hard to guess if there will be any offsets to those gains. Without any offsets, the GDP estimate is likely to be raised from 2.2% in last month's estimate to around 2.4%.
