Saturday, August 30, 2014
Weekly Indicators for August 25 - 29 at XE.com
- by New Deal democrat
My Weekly Indicators post is up at XE.com.
This was one of the most universally positive weeks I have ever seen. Only one series was negative and one neutral. Everything else was positive, several strongly so.
Friday, August 29, 2014
Comprehensive housing summary for July 2014: prices peak, but lower interest rates are putting an end to the sale slowdown
Wtih yesterday's report on pending home sales, housing data from the first 7 months of 2014 is in the books. The data clearly shows that earlier this year was the trough of the housing slowdown, and that lower mortgage rates - down over .5% from the beginning of this year - are beginning to have a positive impact. At the same time, it appears that housing prices hit an interim peak earlier this summer, and are stabilizing if not in a slight decline.
I've seen some very poor analysis of the housing market, that I won't link to, in the last month. For example, I've seen an argument, unsupported by actual data, that prices or even inventory, rather than interest rates, lead sales. This is demonstrably false if you simply look at the data.
You really need to remember the following. The housing market tends to cycle in a regular order:
- 1st, interest rates turn
- 2nd, permits, starts, and sales turn
- 3rd, prices turn
- 4th, inventory turns
Because of the time lag, prices and inventory may still be reacting to a move in interest rates that has since reversed - and that appears to be the case now. July is when the turn in rates and prices became clear.
Interest rates
First, here is a graph, covering the last 30 years, of the YoY% mortgage rates (inverted so that higher rates give a lower value, blue) vs. housing permits, YoY change in 100,000's (red):
Interest rates on mortgages went up from 3.4% in early May 2013 to a high of 3.6% in August of last year. On 16 of 19 occasions since the end of World War 2, that big a change led to a YoY decline of at least -100,000 in permits. In this case, housing permits drifted back lower, down to 4.1% at the end of June of this year, and in the last month have been on average about -0.3% lower than they were at this time last year.
The YoY decline in interest rates suggested that we would start to see some improvement in permits, sales, and starts, although probably muted since rates have not returned to 2013 lows. In July, in two of the three series, we did.
Interest rates
First, here is a graph, covering the last 30 years, of the YoY% mortgage rates (inverted so that higher rates give a lower value, blue) vs. housing permits, YoY change in 100,000's (red):
Here's a close-up of the last 5 years through July:
Interest rates on mortgages went up from 3.4% in early May 2013 to a high of 3.6% in August of last year. On 16 of 19 occasions since the end of World War 2, that big a change led to a YoY decline of at least -100,000 in permits. In this case, housing permits drifted back lower, down to 4.1% at the end of June of this year, and in the last month have been on average about -0.3% lower than they were at this time last year.
The YoY decline in interest rates suggested that we would start to see some improvement in permits, sales, and starts, although probably muted since rates have not returned to 2013 lows. In July, in two of the three series, we did.
Permits, starts, and sales
Here is a graph of the change, in thousands, YoY of starts (blue), permits (red), new home sales (green), and existing home sales (orange) (note that the St. Louis FRED does not track pending home sales):
Both of these graphs show the clear deceleration in the housing market through 2013 and into outright declines in the early part of this year. New and existing home sales have been consistently negative YoY, and permits ended up at midyear only +2% in the first half of 2014 compared with the first half of 2013.
But with July's data, we can see that the trough of the housing slowdown is in place, and that there has been a significant positive turn in the last several months. Only pending sales were negative YoY by -2.1%, although they were up month over month for the fifth month in a row.
In summary, through July 2014:
- Permits are down -0.9% from their October 2013 high, but up +12.6% from their January 2014 low
- Starts are down -1.1% from their November 2013 high, but up +21.9% from their Januay 2014 low
- New home sales are down -9.8% from their January 2013 high, but up +2.2% from their March 2014 low
- Existing home sales are down -4.3% from their July 2013 high, but up +12.2% from their March 2014 low
- Pending home sales are down -2.1% from their June 2013 high, but up +12.4% from their February 2014 low
The impact of demographics on permits, starts, and sales
I suspect the situation this year is analogous to the late 1960's (one of the four exceptions to the rule that rising interest rates cause an actual decrease in sales), when Boomers first reached adulthood and the existing apartment stock was nowhere near adequate to the task. Multi-unit starts skyrocketed, despite higher interest rates, while single family homes languished. It was an era of generally rising interest rates, and any temporary decline in interest rates was met with heightened housing activity.
Now it is Millennials. Now as then, it is only multi-unit (apartment) construction that is carrying the recovery in housing this year. Even with the overall July increase, single family home starts and sales have completely stalled. Here is a graph of the YoY% change in single family house permits (blue) and multi-unit permits (red) since the beginning of 2011:
Prices
The Case Shiller 20 city index for July showed an actual decline from May and June. By this important measure, prices have actually already made an interim peak:
It is likely that we have seen an interim, seasonally adjusted price peak in housing. It should decline for awhile before bottoming, as sales have already bottomed.
Inventory
With housing prices still increasing YoY, even if they are near or at or slightly past their seasonally adjusted peak, we would expect to find more inventory entering the market, as potential sellers hope to take advantage of the improved pricing situation. And that's exactly what we find. Below is the graph of combined new and existing home inventories:
The inventory of houses for sale is not just increasing, but it is increasing at an accelerating rate YoY.
In summary, through July 2014:
- 1. Lower interest rates have revived sales from their trough earlier this year. At the same time, because interest rates are still higher than they were several years ago, I am not expecting a renewed boom in sales.
- 2. As shown by the Case Shiller Index, prices hit an interim peak earlier this summer. Because the lower interest rates are already feeding through the system, at this point I am expecting only a slight decline, or more generally a period of stable prices.
As I have noted a number of times in the last month, the biggest issue in the second half of this year is how much interest rates go down, and whether the decline is transitory or persists. New housing sales are very important to the economy, and tend to feed through the entire economy over the course of a year or more. For now, the trend is in the right direction.
Thursday, August 28, 2014
2Q deflated corporate profits, real residential spending add to caution about 2015
- by New Deal democrat
I have a new post up at XE.com, discussing several important datapoints from this monring's revisions to second quarter GDP.
The remainder of 2014 still looks good, but there is at least one additional reason to remain cautious about 2015.
No matter how you measure, wages have stagnated
- by New Deal democrat
We have a variety of economic data series to track wages, including measures of average wages, median wages, and wages per unit production hour. There are at least 7 such measures.
In addition to the monthly average hourly pay report (listed first below), there are 4 quartely series:
- The most commonly known measure is that of average hourly pay for nonsupervisory workers, which is part of the monthly jobs report.
- The Bureau of Labor Statistics, which conducts the household employment survey, also reports "usual weekly earnings" for full time workers each quarter.
- The BLS also measures the Employment Cost Index quarterly.
- The BLS also measures "business sector real compensation per hour" quarterly.
- Another quarterly measure is unit labor costs, which measure how much labor gets paid per unit of output.
There are also two annual series:
- The BLS reports occupational employment statistics which are reported annually.
- Additionally, the Social Security Administration measures annual net compensation from actual W-9 tax withholding forms, but this has only been issued through 2012. This method means that the result is subject to change based on the total hours worked (remember that in the recession we lost 6% of jobs, but almost 10% of aggregate hours).
The first graph, below, tracks monthly average (mean, not median) hourly wages (green), median wages from the employment cost index (red), and unit labor production costs (blue). All are adjusted for inflation using the CPI. Since the quarterly index of median wages only started in Q1 2000, I have normed the indexes to 100 at that time:
As you can see, real average wages have risen fitfully, mainly reacting to the price of gas). Median real wages have essentially been flat for 10 years. This is certainly not what I would call good, given all the productivity gains during that same period. You need consumers to have the money to buy, before real growth in selling picks up as well.
Speaking of productivity gains, here are unit labor costs (i.e., how much labor does it cost to produce one widget) through the second quarter of 2014:
But since the price producers have charged for widgets has changed over time (in other words, is it really a labor gain if labor costs 2% more per widget, but the producer charges consumers 4% more per widget?), here are real unit labor costs adjusted by inflation:
This tells quite a different story. The share of payments to labor has not kept up with the real price of production -- which is the flip side of record corporate profits.
The remaining 4 measures do not come with graphic presentations either by the BLS or the St. Louis FRED. So I will provide them in a table, below.
Two of the remaining series uses tax return data. Once a year, the Social Security Administration reports on the median salary of those who pay into it via their withholding taxes. Secondly, every year the BLS reports on mean and median hourly wage rates for a panoplay of "occupational employment," and also give the average and the median for the aggregate sample. The most recent report, from this April, reported on the data as of May 2013. The below table gives the median wage, adjusted by the CPI.
A third measure in the table below is the quarterly measure of "usual weekly earnings" of wage and salary workers. That report only goes back to 2004.
To reiterate, all of these are adjusted for inflation:
| Year | Social Security | Usual weekly earnings | Occupational Employment | ||
|---|---|---|---|---|---|
| 1999 | 27,164 | --- | --- | ||
| 2000 | 27,374 | --- | --- | ||
| 2001 | 27,713 | --- | 16.78 | ||
| 2002 | 27,784 | --- | 16.95 | ||
| 2003 | 27,657 | --- | 16.91 | ||
| 2004 | 27,903 | 332 | 16.80 | ||
| 2005 | 27,331 | 333 | 16.71 | ||
| 2006 | 27,832 | 335 | 16.80 | ||
| 2007 | 27,984 | 345 | 16.70 | ||
| 2008 | 27,468 | 342 | 17.22 | ||
| 2009 | 27,584 | 337 | 17.12 | ||
| 2010 | 27,382 | 335 | 17.13 | ||
| 2011 | 26,963 | 337 | 16.86 | ||
| 2012 | 27,413 | 335 | 16.71 | ||
| 2013 | --- | 333 | 16.63 |
While two of the above 3 measues are annual, "usual weekly earnings" is updated quarterly, so here is a more detailed look at that measure through the second quarter of this year:
2013 Q1 332
2013 Q2 334
2013 Q3 333
2013 Q4 334
2014 Q1 336
2014 Q2 330
If you are keeping score, among the median measures, that's a decline from peak to trough is -3.6% for Social Security, -4.3% for usual weekly wages, -3.4% for Occupational employment, and -2.6% for the Employment Cost Index. Average hourly wages declined - 2.6% from 2009 to their trough in 2012. The big, nearly 2% decline in second quarter 2014 of usual weekly earnings is certainly surprising. Since it goes against all of the other monthly and quarterly series, I am inclined to think it is just a case of an outlier panel. We'll see when 3rd quarter results are posted in October.
The bottom line is that, since the turn of the Millenium, real median wages - no matter how measured - have stagnated. Gas prices caused them to rise during the recession, and then decline thereafter as the effects of those gas prices filtered through the economy.
The jury is out on whether that decline has ended. Some measures suggest that real wages bottomed in late 2012 or early 2013. Others suggest that the decline may have continued at least slightly since then.
No matter which measure you use, the fact is that American labor is still not sharing in the fruits of its productivity.
Wednesday, August 27, 2014
Sunday, August 24, 2014
A thought for Sunday: it is time for Judge Griesa to cease and desist
- by New Deal democrat
It's Sunday, so I get to stray from posting economic data, and editorialize....
As most readers of this blog are probably aware, there is contentious litigation in the Southern District of New York Federal Court between several "vulture" funds, and the country of Argentina. In a nutshell, when Argentina was forced to unwind its 1990s currency peg to the $US in 2001, it swapped old bonds for new bonds, but agreed to the jurisdiction of US courts in any litigation over the new bonds. Over 90% of the old bondholders eventually went along, but some bondholders sold their rights to "vulture" funds which insisted on being paid in full. In the ensuing US litigation, the US District Court in New York City agreed with the vulture funds.
The 83 year old senior judge handling the case, Judge Thomas Griesa, however, has gone well beyond making a legal ruling as to who owes whom how much. He has issued injunctions ordering banks anywhere in the world not to allow payments by Argentina on any debt unless it pays the vulture funds in full.
This egregious extra-territorial extension of claimed jurisdiction is bad enough, but if the following statement reported by Reuters on Thursday is true, then in my opinion Griesa has finally stepped way over the line separating Judicial resolution of litigation, and the foreign policy of the United States:
"Griesa said proposed legislation announced on Tuesday by President Cristina Fernandez would violate orders he imposed favoring creditors who refused to accept restructured bonds following the country's 2002 default on $100 billion in debt.""It is illegal, and the court directs that it cannot be carried out," Griesa said at a hearing in New York."
I don't remember Judge Griesa being made the King of Argentina. Let's be clear here: the new law passed by Argentina may violate the terms of its bond swap, and it may violate the orders of Judge Griesa in the US. But for a US judge to be claim that a law passed by another country is illegal, and may not be obeyed, is a breathtaking usurpation of executive authority, and an imperial arrogation of Sovereignty over a foreign country.
To understand how outrageous this is, swap out Argentina for a country with more power. Imagine a US judge Ordering that a law enacted by China, or Russia, or the EU, is "illegal" and may not be obeyed. To say that the blowback against the US would be significant would be an understatement. The blunt truth is, the only reason Griesa can take this position is that Argentina doesn't have the military power to extract a price. Doesn't that sound like a decision that the Executive, not a low level judge, should be making?
I also don't buy that "Argentina" has made its bed, and now must sleep in it. A corporation, faced with a similar issue, could simply wind up operations and its shareholder start a new corporation from scratch. Countries can't do that. And Argentina in 2015 is literally a different population from Argentina in the 1990s or 2001. Demographically, its median age is 30, meaning an absolute majority of Argentinians were not of voting age in 2001, let along the 1990s, and about 15% of its population wasn't even born in 2001! So the coercion being inflicted by Judge Griesa falls largely upon individuals who cannot bear any moral responsibility for the decisions made by former governments.
Judge Griesa has done quite enough damage already. He has now vastly overstepped the proper authority of a US court, by effectively declaring himself the omnipotent Sovereign over another country. It is time for him to cease and desist, or to be made to cease and desist by the Obama Administration.