Approx. reading time 4 min 30 secs
There’s been buzz of a recession developing around the corner.
Alex Rosenberg, a high profile CNBC reporter released an interview with Marc Faber on 18th June 2015, where Faber proclaims a recession is due by the end of 2015. Faber is of course the publisher of the ‘Gloom Boom & Doom Report’ newsletter, so a pinch of salt is required, however his claims are that: the cost of living in the US is rising and corporate taxes, as a % of GDP, are flat.
Further advocates to a USA recession include JP Morgan’s Garrett T Fish, an investment manager for JPMorgan American Investment Trust plc (source) and Raoul Pal, the writer of The Global Macro Investor newsletter, who both believe we’re hitting the seven to eight year business cycle and are due a recession in 2015/2016 (source).
Critics of an incoming recession cite the Baltic Dry Index, which shows the price movement of shipments by sea, hitting a nine month high. Total non-farm employment has increased on average by 211k a month this year (source) and personal consumption expenditure has increased on average +0.38% month to month this year (source).
It’s hard to place an argument for a recession when key figures on employment and expenditure are doing so well.
However let’s take a step back. Firstly it is required to define a US recession and the factors that go into determining it’s official beginning. In the US a recession is called by the NBER Business Cycle Dating Committee. This committee base their decision on four main indicators: Non-farm employment, Industrial production, Real retail sales, Real personal income (others also, but not noted in this article). Therefore for a recession to begin in the US it has to be announced, as opposed to the UK definition of two quarters of negative GDP growth.
Non-Farm Payroll Employment
Definition: An economic indicator reported by the U.S. Bureau of Labor Statistics. Reports the total number of paid U.S. workers of any business, excluding general government, private household, nonprofits and farm employees (source).
Data from the 7th of August show non-farm jobs in July increasing by 215k (source). We have also seen an average monthly increase of 211k (source) so far this year. This shows strength in the labour market in providing employment for individuals.
The graph below helps visualise the correlation between non-farm monthly changes and recessions.
As we can see non-farm payrolls can provide a useful tool in the indication of a recession, as negative monthly % changes tend to be correlated with a recession.
The strength of the current non-farm payrolls shows one positive towards the unlikelihood of a recession occurring soon. However we must understand that the data does not take into account population growth. America has seen an increase in its working population (20-69 yrs old) of 9.1mn between the years 2007 and 2012 (source). Furthermore the natural rate of unemployment is 5.4% (long-term) as of Q4 2014 (source) and with current unemployment at 5.7% on Jan 2015 (source) we can also argue that the employment rates are high, as the economy is adjusting to a recovery from such a severe previous recession. This does not mean the economy is necessarily strong, or recession-proof, it is just a return to normality.
Definition: An economic indicator produced by the Federal Reserve Board. Indicates the amount of output from manufacturing, mining, electric and gas industries. The indicator is currently indexed to the year 2012 and sits at the level 107.5 (source).
The Federal reserve realised data for July industrial production on August 14, 2015 (source). They show that IP increased 0.6% in July, following a 0.1% increase in June. Manufacturing output increased due to an increase in motor vehicle assemblies, mining increase +0.2% and utilities fell -1.0%. The current level of the IP index at 107.5 shows a +1.3% y-y increase.
These statistics are positive for the economy, but the figure for IP has remained rather stagnant since the beginning of the year, when it was at 107.4. Furthermore some critics see IP as a weak economic indicator as it does not take into account inflation and is often subjected to major revisions – a result of the network of underlying indices which create the IP.
Real Retail Sales
Definition: An economic indicator released by the Census Bureau and the Department of Commerce. It measures the sale of retail goods over a stated time period, through extrapolated data sampling. Auto sales are sometimes excluded, as their high price can add volatility to the index (source).
Recent figures show a rebound in July, with a rise of +0.6% and a revision to the June figure from -0.3% to 0.0%. These figures have made the news, as they further add to the case that the Fed will raise interest rates this year.
However if you observe the graph below you notice that real retail sales are still hovering below the regression line that indicates variation from the long term trend.
Although sitting below the regression does not necessarily indicate recessionary tendencies it does feed the argument that retail sales aren’t as strong as the media suggests. Furthermore we see a dip in the beginning of 2015, which is accredited to a severe winter – but further volatile movements suggest it is not as a straight forward excuse as the similar fall in January 2014.
We can deduce that real retail sales are nearly $6,692 mn above the pre-2007 financial crisis and recession, so there is clear growth (source).
Real Personal Income (excl. current transfer receipts)
Definition: Released by the Bureau of Economic Analysis it is an indicator of the amount of income received from salaries, wages, dividends, interest, rental income etc. taking into account the effects of inflation on purchasing power (source). The exclusion of current transfer receipts means that ‘benefits received for no direct services are excluded from the figures; e.g. medicare, unemployment assistance, social security’ (source).
In June we saw a rise of +0.47% and there is a current y-y gain of +3.7%. The graph below shows how it has grown over the past 55 years.
As you can see the graph is not the clearest indicator of an economic recession, as the figures tend not to deviate too much from the long-term trend during a recession. We do, however, see a large drop towards the end of the 2008 recession.
Although correlation does not mean causation the graph puts a strong case against a recession happening soon due to the growth since the last recession. N.B. the peak in 2004 is due to a one off dividend from Microsoft and the peak in late 2012 is due to year end tax planning strategies being implemented.
The graph below shows the combination of factors, highlighting the previous recessions.
As the graph shows the indicators all demonstrate a correlation with recessionary periods, it further shows that it is taking a while for the indicators to reach their peaks following the 2007/2008 crisis, however this is testament to how far they fell during the previous recession.
With the strength of these indicators demonstrated, and considering they are the most important four indicators for the NBER in declaring a recession, I’d be happy to say the USA is not due another recession for at least another year. However I would say that none of these indicators are perfect, as we saw from the analysis non-farm employment suffers a lack of adjustment to population growth and industrial production is subject to many underlying indices that can make it volatile and unreliable. They do however provide a good basis for analysis, and their current 2015 positive growth would indicate the USA is safe for now.