WINZ Misusing the Otago Food Cost Survey

The Department of Human Nutrition of the University of Otago produces a yearly food costs survey. It’s intended to provide a statistical measure by which the cost of living can be inferred. As a tool for showing trends in food costs over time, it’s invaluable. However, Work and Income offices have been using this survey as a yardstick for how much should be given to those applying for food grants.

This is reflected in the methods used by the university, packaged items surveyed are suitable for a small family, so provide a steady benchmark for the costs which could be typically expected. The cost per gram is then calculated and this is multiplied by the average amount each person should be expected to consume in a week.

Shelf prices only are used (not specials/promotions). If more than one brand is available, then the lowest priced item (including in-house brands) is recorded. The package sizes surveyed are appropriate for a family of four. If the specified package size is not available, then the price and size for the next closest size is recorded. Some produce items such as broccoli and avocados are sold per item rather than weight. For these foods, five to six items are weighed and the average is used.

Survey averages and actual costs can differ significantly. A family of 4 might be expected to consume 840g of cheese in a week, so the cheese surveyed is in 1kg blocks. A 1kg block of cheese currently costs around $8.50, so the survey assumes the cost per gram of cheese is $8.50/1000g = $0.0085/g and the cost calculated by the survey for the same family of 4 is 840g x $0.0085/g = $7.14. For a single woman, the cost would be 210g x  $0.0085/g = $1.79.

These costs clearly do not reflect the reality for a single woman. Perhaps, on average, over the year she’d spend $1.79/week on cheese. However, if her pantry is empty, she’d more than likely spend $6.20 on a 500g block.

The practices of these Work and Income service centres do not reflect the differences between the survey and reality, somebody applying for a food grant will have almost completely empty cupboards. In order to be given a food grant, you need to have an immediate and essential need for food, which means you must have almost none and no money with which to purchase more. This makes application of the Food Cost Survey entirely inappropriate, its usage is not at all reflective of its purpose, and it falls flat as a result.

To make matters worse, Case Managers often like to perform a further calculation to reduce the amount of food grant given to the applicant. They will multiply the amount supplied by the survey by the proportion of the week remaining until the next benefit payday. A solo mum with a 5 year old would be expected to pay an average of $96/week for food at the absolute minimum, so if there are 4 days remaining until next payday, the food grant amount will be calculated as $96 x 4/7 = $54.86. This is clearly ridiculous, as it assumes the two would buy 4.6 eggs, 2.6L of milk, 120g of margarine, 160g of sugar, and 57g of tomato sauce.

Additionally, the Food Cost Survey has 3 different bands which can be used to estimate food costs; basic, moderate, and liberal. Work and Income consistently use the basic band, which assumes that all food is cooked from scratch, and doesn’t allow for any take-away or pre-prepared meals.

Further, the section of the Food Cost Survey covers standard ingredients costs, but not all the ingredients required to make a meal. A family so short on money that they have to apply for a food grant likely won’t have enough dishwashing liquid or seasoning for food. They might also lack the cooking facilities required to produce a decent home-cooked meal.

There is only one person who knows how much money is needed for a food grant, and that’s the person applying. The questions I always ask when doing advocacy are “how much money do you spend on food for a week?” and “how much money do you need?”. These are really the only questions Work and Income should ask when somebody comes in to apply for a food grant. The costs could be more or less than what they calculate from the survey, but they will accurately reflect the real need of the applicant.

Clearly, Work and Income have misinterpreted and are misapplying a good piece of academic research. The Food Cost Survey is fantastic for showing historical trends of food costs, and shows how these differ by region and age/sex demographic. It might have a place, if correctly interpreted, to set a benchmark for the absolute minimum that is allowed to be given out in food grants; but it is currently being abused as a tool, causing further suffering for those who need the most help.

Leave a Reply

Your email address will not be published. Required fields are marked *