Inflation miscalculated?
21/07/2008 08:37  - (SA)  

Want to know more? Answerit can help.

Inflation - 'SA needs to know' Mboweni lashes out at Investec

Cape Town - The two-year delay by Statistics SA (StatsSA) in implementing the rebasing and reweighting of the inflation basket is having serious implications for the economy.

According to a report published on Fin24.com, calculations by Investec Asset Management have shown that the real inflation rate in the economy is probably far lower than the official inflation number, echoing the predicament in 2003 when CPIX inflation was found to have been overstated by 1.9%.

News24 spoke to Tony Twine, senior economist at Econometrix, to find out what this means for the average consumer.

This article at a glance:

Does this mean that we are likely to see things slowly return to normal in the next few months? Or has the worst yet to come?

News24: What do you make of Tito Mboweni's assertion that Investec stood to benefit from the announcement?

The Governor has a point. The banks are all involved, perhaps independently, in a heavy weight propaganda onslaught against any further monetary policy tightening by the SARB. Hence all the talk about a looming recession, all of which is emanating from the banking economists. There is nothing wrong with this, as long as players in the economy are aware, and do not get scared into making it a self-fulfilling prophesy.

News24: Can you explain why we are now hearing inflation, or CPIX, might not be as high as we thought?

CPIX inflation is the rate of change of the price level of a given basket of household goods and services, as calculated for the Consumer Price Index basket of all household spending items excluding mortgage bond interest payments.

Obviously, the rate of change of the price level of the basket will depend on what is included in the basket, and in what proportions. If we alter the contents or the proportions of the contents, the value of the index will change, as will the rate of increase of the value of the index.

Every five years, large scale surveys are used to estimate what households spend their money on, and in what proportions. The currently active base year for the CPIX is year 2000.

Recently published base year spending weights (all proportions) are for the new base year, which will be 2005. A shift in spending patterns between the two successive base years has meant that the proportion spent on food has shifted downwards from about 26% in the 2000 base to around 16% in the 2005 base.

Food prices have been one of the inflation leaders of recent years, but the reduction in the proportion means that their impact on inflation, being the product of their weight x their inflation rate, will be smaller when the 2005 basket is used than it was in the 2000 basket.

The 2005 basket will come into effect for the first time when the CPIX for January 2009 is calculated and published, which will be around the last Wednesday of February 2009.

News24: Why must the method of calculating inflation be changed in such a way?

The change is simply an update of the weights for each of the hundreds of items, grouping upwards into 18 main groups (e.g. food, transport, clothing and footwear, household costs), which is conventionally executed every five years.

The fuss at the moment is caused by the considerable delay in implementing the 2005 base, which would have usually been put into effect before the end of 2007.

StatsSA have clearly had problems in executing the base transition, which is a huge job, probably because of skills and employment level constraints, just like other government departments.

News24: In simple terms, how is inflation calculated, and what is this "basket of goods" we keep hearing about?

For each base year, household spending is measured by a large sample (around 30 000 respondent households) research study, which measures household spending in the particular base year of the index. It may be found that, for households as a whole, carrots account for 0.3% of household spending, bread 1.1%, men and boys clothing 2.5%, women's clothing 3.2%, petrol 5%, car repairs 0.5%, and so on through a list of over 200 items. For an item to be included in the index, households must spend a minimum of 0.01% of their total spending during the base year on it.

Individual line items are then aggregated into sub-groups. In the fictitious example above, carrots and bread would combine to contribute 1.4% to the food basket within the CPIX basket. Other food items will bring this share up to around 16% in the case of 2005, which would have been closer to 26% as food's share in 2000.

The clothing items mentioned in the fictitious example would aggregate to a contribution of 5.7%, contributing together with other items to the clothing and footwear index within the CPIX.

Petrol and car repairs would aggregate to 5.5% share of the CPIX basket for transport, which would also include spending contributions from other transport items ranging from bus fares to toll road charges.

Each of the groups, as well as the total for CPIX, can then be given a price value each time that its prices are measured (conventionally monthly).

These prices are usually expressed in index form, where the average price for each and every sub-group and the basket as a whole is set to a value of 100 in the base year. As the price index changes over time, the rate of change from one point in time to another (usually from a given month against the same month of the previous year) can be calculated, which is termed the rate of inflation.

News24: What are the most important changes to this new "basket of goods"?

The proportions, or weights, of the constituent parts of the basket. The weight of food in the basket has already been mentioned, but all other weights will have shifted between the two base years, because spending patterns would have changed quite significantly between 2000 (a relatively tough year for consumers) and 2005 (when we were right in the midst of the biggest consumer boom the country has ever seen).

News24: Why were these changes not implemented sooner?

You will have to ask StatsSA, but I suspect that the earlier comment regarding skills and staffing will turn out to be fairly close to the bone.

News24: What is the knock-on effect if inflation is actually lower than we expected? Does this mean consumers have been unfairly punished?

The measure of inflation is used as a general benchmark for many household, business and government policy decisions. Households in the know may base their expectations of future price changes on what they read about the officially calculated inflation rate. Business will do the same, as well as using the CPIX inflation rate as a benchmark in formulating pay increases for their workers.

Many pay increase formulae rest on an approach of "CPIX inflation + another amount", where the other amount attempts to compensate for worker performance, gains in output by the firm, productivity gains, or the opposite of any of these.

The CPIX inflation rate is obviously very important in this approach, because the higher it is, the bigger the pay increases will be, and vice versa. Government will use the inflation rate as an input to its policy decisions, including fiscal management and policy (setting tax and government spending levels and monetary policy, setting interest rates, or the price of credit).

Have consumers been punished? Perhaps, but perhaps not. For instance, interest rates may have risen higher and for longer than would otherwise have been the judgment of the Reserve Bank, but this would have been more than offset by higher than "fair" wage increases, because of the contribution of CPIX inflation to so many wage adjustment calculations.

News24: Will we see rates go up again or have they finally peaked?

This is another question to which the answer is not purely determined by the measure of CPIX inflation. Interest rates do not directly bring down the rate of inflation when they rise, neither do they directly increase inflation when they fall.

The interest rate is really nothing more than the price of credit (not the price of money, which is simply what you can exchange the money for). By manipulating the price of credit, the monetary authorities attempt to change the rate of creation of credit, which impacts on the rate of growth of the money supply.

If the money supply grows much faster than the level of output of real goods and services in the economy, we will end up exchanging more and more money for each unit of real goods and services produced by the economy, i.e. prices will rise, as will the rate of inflation. Slower credit growth would depress the rate of inflation.

It is therefore unlikely that a recalculation of inflation on a new base would lead to a change in Central Bank policy. For that to happen, we will have to wait for the rate of growth of credit to come down sharply from the current annual rate of about 20%, before they relax monetary policy, mainly in the form of interest rates.

In other words, there is no direct linkbetween the inflation rate and the interest rate, but rather an indirect link through most of the rest of the economy outside of these measures.

News24: Even using the new calculation, inflation is still too high. What is causing this and what can we do about it?

The main drivers of the current high level of inflation are food and fuel prices, both of which are being driven upwards by global economic forces.

Other groups of prices within the CPIX are now also rising at rates above the targeted CPIX inflation band of 3 - 6%, which means that the impacts of the exogenous food and fuel prices are feeding through to other prices of goods and services in the consumer basket.

We cannot do anything about the global pressures, because we are too small as individuals or even as a National entity. The only thing that we can do is to try to dampen the spread of the exogenous price increases into other prices, which the SARB is seeking to do by the mechanism outlined above aimed at reducing money supply growth so that the amount of money in the economy per unit of real output is prevented from rising as quickly as we have seen it, thereby combating price rises.

News24: Did StatsSA truly make a mistake or is the argument simply over when these new calculations should have been implemented?

The latter is more true than the former. In rushing a similar process for the Producer Price Index (PPI) basket, mistakes were made, which had to be corrected in terms of published data after its initial release, which was very embarrassing for them.

News24: Does this mean that we are likely to see things slowly return to normal in the next few months? Or is the worst yet to come?

I wonder what "normal" means? The new weights will not be put into effect until the CPIX and its inflation rate are published for January 2009, which will only happen in February 2009. The 2000 base will remain in effect until then, so no technical correction of the kind discussed above will be in place until then.

CPIX inflation is likely to continue on an upward trend at least for June and July (both yet to be published), and possibly for longer. It all depends on the way in which prices behave over the next months, including the big and nasties in the shape of food and fuel prices.

If these continue to be the price inflation leaders into the New Year, then the CPIX inflation rate could correct downwards quite sharply (currently anticipated to be by around two percentage points) in the January 2009 measurement.

By itself, this will not be sufficient to trigger a monetary policy change, but does begin to set the scene for first a sideways move of Interest Rates, and later a relaxation, just as long as credit creation and money supply growth move in the right direction over the rest of 2009.

Tony Twine, of Econometrix, thank you very much for chatting with News24.

- News24

What is this? Yahoo Digg Del.icio.us Facebook Brought to you by OUTsurance Car Insurance
	News24 Headlines on your Facebook profile		News24 on mobile

• Most Read
• Most Commented
• Most Emailed

Clean shave led to rapist
Mom misses son's last call
'ANC must stop mocking Jesus'

Race tension growing - survey
No more nightmares for Courtney
SA uncertain about Motlanthe

'The ANC has lost control'
Inmates' freedom 'worrying'
Balance the blame

ANC no 'exclusive authority'
Get out of my way!
'May the fox be with you'

'May the fox be with you'
High alert at Indian airports
Heavy snowfalls hit UK

Race tension growing - survey
Clerics: ANC claim not blasphemy
New R7bn dam gets the nod