Sun | Sep 23, 2018

Recession talk a bit of a stretch

Published:Sunday | March 1, 2015 | 12:00 AM
Colin Bullock

In the Financial Gleaner dated February 27, 2015, an article written by Aubyn Hill titled 'Low growth hurting young adults' included a section, "The economy slipped back into recession". This section stated that "Jamaica is officially in a recession if the PIOJ estimate holds".

The article further alludes to a definition of a recession as occurring when an economy suffers two consecutive quarters of contraction. In another article in the same Financial Gleaner titled 'The stage is set ...', reference is also made to the economy technically being in a recession following two quarters of decline.

Any inference of a recession from the estimates presented by the Planning Institute of Jamaica (PIOJ) at the quarterly press briefing on February 24, 2014 is a technical misrepresentation and will serve to misinform the public.

The standard technical approach sees a recession occurring when there are two consecutive quarters where seasonally adjusted output is lower than in the previous quarter. Thus, for example, for three quarters ending in June, September and December, a recession may be said to occur if the seasonally adjusted output in the quarter ending September was lower than that for the quarter ending June, followed immediately by even lower seasonally adjusted output in the quarter ending December.

The data used are for sequential quarters, and seasonal adjustment is used to correct for the reality that economic activity may, for various reasons, fluctuate throughout the year. For example, (a) the ending of a sugar crop will lead to lower sugar manufacture in the quarter following crop-over, (b) consumer spending peaks in Christmas and subsides thereafter, and (c) housing rentals for university students decline in summer, and so on.

The preliminary data presented by PIOJ fail the test of capability to measure recession. First, it does not compare sequential quarters but rather offers a comparison with the corresponding quarters of the previous year.




Second, this year-on-year comparison is done because the preliminary data are not seasonally adjusted. The comparison with the corresponding quarter of the previous year seeks to minimise issues of seasonal fluctuations on the expectation that seasonal factors tend to be similar in the same quarters of different years.

The PIOJ's preliminary estimates are to be interpreted as meaning that output in the September and December 2014 quarters was respectively lower than in the September and December 2013 quarters. Even setting aside the absence of seasonal adjustment, the data presented by PIOJ at its quarterly press briefing cannot be used to support any inference that output in the September 2014 quarter was lower than that in the June 2014 quarter and that output in the December 2014 quarter was lower than that in the September 2014 quarter.

It is worth noting that the Statistical Institute of Jamaica produces official seasonally unadjusted and adjusted data three months after the reference quarter. It therefore means that the country is currently not able to accurately determine the seasonally adjusted sequential performance for the relevant quarters until March 31, 2015, when seasonally adjusted data for all three quarters ending June, September and December 2014 would be available.

It is regrettable that the articles have misinformed the public by advancing inferences that cannot be supported by data that are currently available. We do not believe that this was the intention of the authors.

n Colin Bullock is director general of the PIOJ.

Email feedback to