My Respond to Bawumia

I have resisted many requests for radio interviews on this matter (partly because I thought he might have been misquoted), but having read the original article attributed to him, I am at a loss as to what might have prompted such a gratuitous attack on institutions whose data he used in the past and which he continues to use (even if selectively and deceptively so). Indeed, in the case of the central bank, he helped generate those statistics.

He bases his criticism of the changes in the forex and inflation rates on his “humble opinion” and on the strength of this opinion concludes that they are “not credible”. Alas, in such serious and contentious matters, opinions – humble or not – have no place at the table. Either you have your facts and techniques right – and truthfully so – or you don’t.

I am sure both GSS and BoG are capable of defending themselves, but as a citizen, I am no less interested in the accuracy of the statistics that these two institutions put out than my friend Dr. Bawumia is, and so I feel compelled to jump into the debate.

Let’s start with inflation, where Dr. Bawumia is reported to have said the following (and I am assuming that what appears on the web – in this case Myjoyonline.com – is in fact correct):

“Any shopper in Makola in Accra, or markets in Kumasi, Akatsi, Techiman, Cape Coast, Wa, Bunkpurugu, or Bolga markets would laugh at you if you tell them that average prices of food and non-food items have gone up by an average of only 1.5% so far in 2014 and that prices are not increasing as much this year as they were last year (2013).”

Dr. Bawumia himself has criticised such an arbitrary and haphazard approach to inflation discussion in the past, so one wonders why he enlists that now.

And if the GSS is being as dishonest as he would have us believe, then their inflation rates should be going down, not up. But the available evidence shows that that is not the case.

In fact, last year, when the GSS rebased the Consumer Price Index (CPI), which is used to calculate the rate of consumer inflation, they reported instances where the new index actually showed a HIGHER (and implicitly a more accurate measure) of inflation than the old index did. Hence, the May 2013 inflation rate based on the new CPI, for example, was 11.1%, instead of the 10.9% that would have been recorded under the old CPI. The inflation rate has consistently risen since then, reaching 15.3% in July 2014. (Because this is an average, some components are obviously much higher, as I will show below).

Even more worrying is the fact that the good doc nominates some markets with no obvious logical basis and suggests that observed prices in these markets somehow trump the validity of the inflation figures provided by the GSS. Of course, anybody can come up with their own wishful list of markets, which would then lead to a lot of confusion, hence the need for a commonly agreed metric – the CPI. We can’t pick and choose at will. We must be governed by some rules, at least.

We must also be mindful that the CPI (and the inflation rate derived thereof) is an average of a range of representative household products, whose prices rise, fall or remain the same in varying proportions across time and space. Some changes are almost instantaneous while others work with a lag, all of which explains why we have different inflation rates for the 10 regions, while utility price hikes hit some sectors immediately but take time to work their way into others. The professional definition of inflation as the GENERAL increases in prices, and not necessarily individual prices in some haphazardly chosen markets, is obviously not without basis.

In the case of Dr. Bawumia, he derives an “average of only 1.5%” inflation from GSS figures for 2014, but the CPI available at the GSS’s website does not support that conclusion. The following are the indices for calculating selected inflation rates for the year.

July 2013: 113.6
December 2013: 116.6
July 2014: 131.0

Out of them we derive the July 2014 year-on-year inflation rate (which is 15.3%) and the July 2014 year-to-date inflation (which is 12.3%).

Of even greater and broader importance are the disaggregated sub-components of these averages; they provide more intuitive and realistic information about market conditions than do headline inflation figures. And these can be found in the monthly inflation newsletter that the GSS dutifully publishes but is seldom discussed. Any serious analysis of inflation must, therefore, reflect the content of this important newsletter which breaks down the headline inflation into what we really experience in the market.

Dr. Bawumia doesn’t do that. Instead, he dwells on “reported inflation numbers”, which in his view “suggests that something is not right”. He adds: “The rate of increase of prices in 2014 by our collective experience is clearly the highest in recent times, underpinned by the 40% depreciation of the exchange (sic) so far this year, increases in petroleum, utility and transportation prices. The prices of most goods and services have virtually doubled and continue to increase daily”.

He then drops a bombshell (or so he thought), which is worth a full quote because it goes – falsely – to the heart of his argument: “This notwithstanding, the Ghana Statistical service (sic) data on inflation suggests that price increases in 2014 are slower than they were last year (2013). Between January and July 2014 inflation increased by only 1.5% (13.8% to 15.3%) after all the increases in petroleum, utilities and other prices we have witnessed. For the same period in 2013, inflation increased by 1.7%. This means that we were experiencing marginally faster price increases in 2013 than in 2014. Unfortunately this data does not capture our collective experience. Almost every individual, household, business would tell you that prices are increasing much faster this year than has been observed in many years. So why is the Statistical Service data not capturing this?”

This is a most dubious analysis of inflation; no wonder it fails to “capture” our so-called “collective experience”, which the GSS lays out in unambiguous statistics on its website for anyone who cares to know. In the June 2014 CPI newsletter, for example, the GSS reported a year-on-year inflation rate of 53.6% (yes, fifty-three point six percent) for the “housing, water, electricity, gas and other fuels” category of the non-food subsector, up from 35.0% in December 2013; the figure for July 2014 was 62%; together, we see a rising trend driven by hikes in utilities prices prompted by government’s policy to scale back subsidies.

It is clear from the foregoing that both at the aggregate and sub-component levels, the Statistical Service has been faithful to the realities of the market and the sentiments of the public, and that it is Dr. Bawumia rather who has sought to confuse us with a tortuous computation of inflation that bears no resemblance to what Ghanaians are actually experiencing.

Surely, we deserve an apology for such an insult to our intelligence.

(By the way, an increase in inflation from 10% in January to 15% in June of one year (a 5-point increase) and from 30% to 32% between the same months in a second year (a 2-point increase) does not necessarily mean that the statistics authorities, whoever they may be, are lying and that life is somehow better in the second year than the previous one; prices are still rising at a faster pace (between those two months, at least) in the second year than they did in the first – some 17 points higher in June.).

I must also briefly address the matter of the cedi’s depreciation. I am sure the BoG is more than capable of defending the integrity of its data; my interest is the computation, no matter the source.

No one, of course, doubts the gravity of the cedi situation, and much is being done by government to address it. Even a 10% depreciation is too high and must be cause for worry – and it is. But that's no excuse for people to take liberties with facts and reasoning and confuse a generally unsuspecting public.

In the example cited by Dr. Bawumia, we seem to have a case of crossed-wire analysis and its associated faulty conclusions. He states: “At the end of December 2013 the cedi was GHC2.2 per US dollar. Today, it is averaging some GHC 3.8 on the interbank and forex bureau markets. This represents a depreciation of over 40%.”

But the GHc2.2 is the central bank’s reported interbank rate, whose integrity he doubts, while the “averaging” of the interbank and bureau rates are Dr. Bawumia’s own contrivance, which of course he believes in, a clear case of wilful crossing of wires to produce a predetermined result. Either data from the Bank or his own, but he can’t have it both ways in a selective manner.

To be sure, there has been some fixation with exchange rate computations in sections of the public recently after some media houses reported, erroneously, that the cedi had lost “nearly” 40% of its value this year; this fixation has compelled some people to go to great lengths to prove this falsehood in reverse.

Financial analyst Sydney Casely-Hayford recently published a piece on the internet titled, “Getting The Numbers Right. Omane Boamah’s Bad Arithmetic”. Hear him out:

“At the close of December 31, 2013 the quoted rate by the Bank was 2.1628=$1. This is the correct figure to use as the opening rate on January 1, 2014. As at 11thAugust 2014 the BoG quoted rate was 3.035=$1.

“The formula to calculate the extent of depreciation is ((2.1628-3.035)/2.1628).

“This calculation makes the rate of depreciation -40.3%.”

Unlike Dr. Bawumia, Mr. Casely-Hayford at least proffers some confidence in the BoG data; his problem, however, is that he makes a common mistake of confusing an appreciation of the US dollar against the cedi with a depreciation of the cedi against the dollar.

His calculation clearly shows a positive and not a negative change of 40.3% (the quantity of cedis has obviously increased); not happy with that – and obviously not knowing what to do – he simply decided to stick a negative sign in front of his result and be done with. Cute but wrong.

You see, the two cedi amounts in his calculation are actually the values of ONE US DOLLAR at two different points in time; a positive percent change between those values, therefore, means an APPRECIATION of the US dollar against the cedi, in this case 40.3%, which explains why the computer produced no negative sign and Mr. Casely-Hayford had to invent one, brazenly.

To get the cedi’s depreciation (or appreciation), he should have first found the value of ONE GHANA CEDI against the US dollar at two different points in time, which would be the following: 1 divided by 2.1628 and 1 divided by 3.035. If there’s a depreciation, the computer would automatically insert a negative sign (there would be no need to fudge); otherwise, it would be an appreciation. In this case, based on his data, the depreciation rate of the cedi against the dollar would actually be -28.7%, and not 40.3% (which, again, is the appreciation rate of the dollar against the cedi).

I do not by this unwittingly long response seek to diminish or deny the current economic challenges that we are facing as a nation. I only want us to discuss and confront these challenges truthfully and accurately. In that sense, I agree with my friend and brother, Dr. Bawumia, whom I have known for some 20 odd years (and whose intellect I respect, even if I disagree with his analysis), when he says in his piece published on Myjoyonline.com:

“We need reliable data to manage the economy prudently. If the data is wrong, important policy decisions based on such data would be wrong. It is therefore important that when we observe such situations we point them out so that corrective actions can be taken to save all of us from suffering the adverse effects of wrong policy decisions based on such unreliable data.”

To which I say, Amen!

PS: Anyone interested in the numbers behind the monthly headline inflation figures can always access www.statsghana.gov.gh and click the CPI icon. Be sure to download the Powerpoint presentation on the rebasing of the CPI last year; quite informative.
:: Ghana Statistical Service ::
statsghana.gov.gh