Last Friday saw the release of the latest unemployment and employment statistics from the US Bureau of Labor Statistics. These were eagerly awaited because the recent theme for the US economy has been that it is growing but not by enough to make much of a dent in the unemployment figures. We keep getting weekly initial claims figures on a Thursday which suggest an improvement in the employment situation but not by enough to help with unemployment as overall it has mostly only been enough to keep pace with population growth. Why does this matter? At this stage of the recession if you compare it to previous post second world war recessions unemployment has been higher for longer than any other. This has led to the phrase “jobless recovery” being used in spite of the fact that it is something of an oxymoron and the jobless are hardly likely to call it a recovery. I am afraid that this phase is one which reminds me of George Orwell’s phrase “Some are more equal than others.”
The Contradictory Nature of the numbers we received
The BLS released these two headline numbers.
The unemployment rate (9.0 percent) declined by 0.4 percentage point for the second month in a row. (See table A-1.) The number of
unemployed persons decreased by about 600,000 in January to 13.9 million, while the labor force was unchanged.
Total nonfarm payroll employment changed little in January (+36,000). Manufacturing and retail trade added jobs over the month, while
employment declined in construction and in transportation and warehousing.
So we had cheers and good humour at the first release quickly followed by frowns at the nonfarm payroll numbers. Those who follow these numbers should have spotted an obvious flaw. How can unemployment fall substantially whilst employment has barely improved? Also if we remember the figures for December they too had a fall of 0.4% in the unemployment rate but there were concerns over the statistical assumptions used. Indeed some had expected a reversal of part of this and so unemployment had been expected to rise to 9.5% which turned out to be another economic forecasting error (these are really piling up over the “credit crunch” period.
This is the point to explain that the numbers are not from the same survey they are from two different ones so whilst in theory they should agree in practice they do not have to. Let me explain how the system works.
US Unemployment and Employment surveys
The starting point is that the numbers here come from 2 different surveys.Fortunately they cover the same time period of a month unlike the UK for example where our unemployment rate is calculated over a different time period from registered unemployment. The unemployment rate comes from the Census Population Survey which surveys 60,000 households and the non-farm payroll numbers come from a separate survey of around 140,000 businesses. This leads to two main thoughts. Firstly whilst statistics can be helpful we are using a sample of 60,000 to tell us about a labour force of around 153 million which has obvious implications for potential flaws and secondly that there is no great reason for the two surveys to be correlated with any precision.
If you look at the two surveys the initial thought is that as nonfarm payrolls has a much wider survey base it is likely to be more accurate. However in practice it is often revised quite substantially. For example this report revised December’s employment gain upwards from 103,000 to 121,000 and November’s employment rise, which was first reported at 39,000 and the revised to 71,000, has been revised again to a total gain of 93,000. Adding these to the January headline figure does bring it more in line with the unemployment numbers, but of course the January numbers are likely to be heavily revised too!
What else happened in these numbers?
I have followed over time the wide measure of unemployment called U-6. This allows for workers being put on part or short time and also allows for what are called workers marginally attached to the labour force. This measure fell from 16.7% to 16.1% which is still high but clearly an improvement. A feature of this number was that part-time employment fell from 8.9 million to 8.4 million which was also welcome.
Not so hopeful was the way that the Labor Force participation rate fell to 64.2% compared to a more normal level of two-thirds. This is the lowest for 30 years and the level has fallen by 2% over this recession. In a labour force of around 150 million this is 3 million people which to my mind puts what might be considered a rather arcane statistic into perspective. And it poses the question where have they gone? It is clear that a large number of people have left the labour force, for example by retiring early.
Long-term unemployment dropped from 6.44 million to 6.21 million which was also hopeful.
There is an obvious problem between the two surveys discussed here. Just to add to the confusion there were some further statistical changes in the numbers which meant that the population of the United States was assumed to have dropped by 347,000 according to the unemployment figures. This further adds to the fog of confusion! We are left with the view that the numbers look like they showed an improvement for unemployment but however you twist them or blame snow we are, in truth, very little wiser about the employment situation.
One matter was clearer however as there were revisions to the numbers going back to 2006. These told us that 8.9 million jobs have been lost in the credit crunch recession of which only 950,000 have been regained. This provides a rather poor background to the figures.
I notice that there has been some criticism of the U-6 unemployment measure showing 16.1% on a seasonally adjusted basis when the raw number was 17.3%. A fair point until you notice that such a divergence appeared to happen at this time last year and came back in line. Only time will tell.
What did markets think? A further surge in long-term interest-rates/ US government bonds fall
The US equity market gave us very little clue as the Dow Jones Industrial Average rose a mere 29 points. However the US Treasury Bond market fell substantially after the news with the price of the long bond ( 30 year maturity ) falling by more than a point to yield 4.73%. This yield has been rising for a while and also impacts on the US 30 year fixed mortgage rate which rose above 5% last week to 5.02%. This is hardly going to help the already struggling US housing market. Shorter-dated yields rose too with the 10 year rising from 3.54% to 3.65% and the 5 year from 2.16% to 2.28%. These are the areas surrounding where the US Federal Reserve has been buying US government bonds which begs the question of how much on a marked to market basis this policy is now losing. When you are the biggest investor in a market which has substantial falls there are obvious problems, not least perhaps that your policy was supposed to make the market rise!
We do not know that the unemployment/employment figures were the only factor on the US Treasury Bond markets mind but by its price falls it seemed to indicate that it was willing to trade on and believe in the improved unemployment numbers.
The implication for UK Mortgage Interest Rates
I wrote this on February 2nd about this situation.
If we look at the two-year the yield rose to 1.4% which compares with 1.18% a week ago and 1.09% a month ago. If we look at the five-year the yield rose to 2.53% which compares with 2.34% a week ago and 2.19% a month ago.
I was referring to our government bond yields and making the point that these yield rises will explicitly affect fixed-rate mortgages. I also feel that they implicitly affect variable-rates ones too. Since I wrote this the underlying situation has continued to deteriorate as out two-year yield has risen to 1.52% and our five-year to 2.66%. Such moves will be putting upward pressure on UK mortgage rates.
These interest-rate rises will take a little time to feed fully into mortgage rates but if sustained they will. I have pointed out in previous articles that the actions of the Bank of England in ending its Special Liquidity Scheme early is likely to reduce the quantity of mortgages available. Now we can see that rises in the price of mortgage finance are happening too. So the mortgage market will be squeezed on both sides and this has an obvious implication for the likely behaviour of house prices in 2011 which look set for falls maybe substantial ones.
Another implication of these interest-rate moves is that they make the official base rate even less relevant as it gets left further and further behind what is happening in the markets. I would remind everyone reading this that this rate is strictly only for overnight and maybe for a month (the time between meetings of the Monetary Policy Committee or MPC). How often do you want to borrow or save or invest for only a month? This provides a background to this weeks debate as to what the Bank of England will do at its meeting this week. My view is that we need a rise and I notice that the Shadow MPC now agrees as it voted for a rise to 1%. Well, it agrees on the direction if not the size!
Can Kicking in the Euro zone: what to do about Portugal
There was a meeting of Euro zone ministers over the weekend and as usual before it there were all sorts of rumours released about what they might do. Again,as usual the meeting has broken up with little achieved apart from a further summit being called for the end of March! So can kicking yes but not very far. Rather curiously I have seen a few articles saying things are under control now in the Euro zone. Those writing this seem to have missed that on Friday Portugal’s ten-year government bond yield was 7.05%, a level leaving her profoundly insolvent. Is that the solution? Also when it was last at such a level we saw a lot of articles in the media on Portugal’s crisis! Is crisis the new solution?
In reality rising levels of the Euro exchange rate and her own government bond yields is a toxic mix for Portugal.