So just how accurate are inflation stats?

What goes up must come down is an idiom that used to be true until the new electricity tariffs came along.

Then the food price got caught in the slipstream. And toll fees are soon to be added to the eye-popping cost of fuel.

The money's gone before the month has started, and that's after cancelling the DStv subscription, along with your Christmas holiday - you're going to need your bonus to pay for switching on your heater next winter. And, by all accounts, things aren't likely to improve in the near term.

The official word is that inflation is at 5%, but when you hand over your credit card at your local grocery store, it feels a whole lot higher than that.

Patrick Kelly, executive manager of price and employment statistics at Statistics SA, says he'll "put his life on it" that the official figure is correct, no matter how much emptier our wallets might feel or how much faster our cash is disappearing.

There are 400 items in the CPI basket, categorised into 12 groups that include food, housing, transport and clothing, right through to funerals and financial services. Each item is given a weight, a percentage of the total, depending on how much the average consumer spends on it in a year. The current weights, which were implemented in 2009, are based on a household survey of expenditure that was done in 2005/6. This weighting is usually revised every five years, but this cycle will be revised in 2013.

Kelly says Stats SA has a policy of over-the-top quality management when it comes to monitoring inflation. He says about 80000 prices are collected from 25 cities around the country each month to keep tabs on percentage changes that would indicate inflationary shifts.

It's no surprise that electricity has the highest rate of inflationary increase - about 20%. However, it remains overall a small percentage of people's budgets - about 2% - and, for that reason, the price increase doesn't distort the overall inflation picture. Even if electricity's percentage of the basket goes up from 2% to 2.5%, reflecting a large increase in price, it still doesn't really change things.

The biggest portion of the CPI is made up of housing costs, which are based on rental prices. In the case of owner-occupied homes, owners are considered to be renting from themselves and the figures reflect this as part of their budget. Supermarket goods count for about 20% of CPI.

Consumers are most aware of the prices of things that are bought regularly, like food and toiletries, rather than high-priced items bought rarely, like cars. They also remember and react when prices go up, but not when they go down or stay flat, which is a trend happening with durable goods like fridges and furniture, and also with computers and cellphones, the prices of which trend downwards after they enter the country.

Kelly says when the electricity price goes up - typically around July - it gets a lot of talk and coverage. This tends to disappear by April, when everyone's factored it into their financial lives, but the routine begins again when the next increase comes into play. Kelly says, by the same principle, the fact that fuel prices are reported on and adjusted each month makes this commodity such a point of contention and discussion.

The problem with power and fuel is that there are no alternatives. One of the ways consumers can get around price increases is by substitution - if beef goes up, they can buy chicken, for example - but power and fuel can't be substituted for a lower-priced commodity.

Kelly agrees that we have not yet seen the second-round effects of the increases, especially in terms of food prices, which have - rather alarmingly - increased less than in other countries, probably because of the strength of the rand.

Economist Mike Schussler doesn't agree with the official statistics. Although he believes the trend is right, he says the absolute inflation figure is probably closer to 6%. He says people are spending far more than 2% of their household budget on electricity and in excess of 10% on municipal services, including rates and water. For lower-income groups, these costs swiftly remove about 25% of their monthly household budget, possibly more than they spend on food, which was weighted at 16% of CPI in 2008.

In addition, toll roads were not factored into the CPI basket, but they now have a significant effect on our transport costs, and if the mooted new tolls are implemented, the impact will increase significantly.

There are several indications that people don't give much credence to the 5% inflation figure. Firstly, few people try to save money in banks, where they are offered, for example, a 5.5% interest rate, even though this should represent a positive rate of return. Secondly, industry is offering 7% increases to employees in wage negotiations, which in all likelihood means they believe inflation is at least at the 6% mark and that what they are offering is above inflation. Workers asking for between 13% and 16% increases are clearly showing their disdain for the official statistics.

Zandile Makgoba, an economist at Econometrix, says employer wage offerings are based on a company's growth and profit prospects, while unions consider living expenses as a basis from which to argue for wage increases.

She says inflation needs to be assessed across the different income quantiles, and that the lower-income quantile is experiencing much greater increases in inflation than consumers in the upper-income quantile. Food and transport are heavily weighted in the baskets of lower-income earners.

Makgoba says consumers can expect an increase in prices in the short term because of the knock-on effects of PPI through factors like electricity and transport cost increases. However, she expects prices to flatten at their new highs towards the end of the year.

There is a glimmer of hope in that interest rates are unlikely to be hiked until people start spending on non-necessary items again. Makgoba says the monetary policy committee will most likely keep rates flat until services inflation begins to rise. Services include non-necessary items like restaurant meals, massages and haircuts.

This will indicate a change in demand for goods, rather than increases in spending as a result of changes to the price of inputs like power and fuel.