NZIER takes issue with Keith Woodford's view that GDP understates the economic contribution of agribusiness to NZ, arguing it's only measuring what GDP is supposed to measure and excludes exports

Keith Woodford's article “Agricultural GDP leaves out more than it includes” raises interesting issues about the strengths and weaknesses of GDP as a measure of economic well-being. We agree with some of the points Professor Woodford makes, but we believe that he is trying to use GDP to do things it was never intended to do. 

There are two aspects to this issue. The first is the difference between costs and benefits of agricultural production. The second is the relationship between the value of exports and the economic benefit of producing those exports.

Starting with the costs and benefits issue, Professor Woodford makes the valid point that the production of goods on farms requires a combination of the productive capability of the land itself, the chosen crop or animals that graze on that land, the managerial skill and effort of the farmer and then many other off-farm inputs. All of these combine to produce the final output.

The economic statistic of GDP is designed to measure, in a single figure with no double counting, the value of all the production undertaken within all markets in an economy in a given period of time.

There are a number of different ways that statistical agencies construct measures of GDP. The one that NZIER used in its recent report on the impact of regulatory reforms on the dairy sector was the production-based measure.[1] Under this approach to calculating GDP, statistical agencies record the final sales of firms in the economy, be it to local consumers or an export. The “final sales” part is important. In many cases, the final producers of a good will buy inputs from other sellers. A farmer will purchase fertiliser from a fertiliser company, veterinary services from the local vet, software, hardware, and the myriad other supplies that go into producing milk on farms. If the statistician simply added up all the sales along that chain, they would exceed the total value of the final sale and give a distorted picture of what is happening.

To get an accurate measure of GDP, statisticians net out the purchases along the supply chain from the final sale price of the goods. This is very similar to how GST works in NZ. When a GST-registered firm sells anything to anyone, it adds GST to the final bill and the consumer pays the GST-inclusive price. However, what the producer returns to Inland Revenue is the amount of GST that they have added to the final selling price, less the GST that they themselves paid for the inputs into their products. This happens right along the supply chain, and the net GST the final consumer pays is the total amount of revenue that the government receives from the entire production process leading up to that final sale.

Production-based GDP works the same way. The statistician records the final value of sales, and then netts off from that amount the cost of all the inputs into the production process. Indeed, GST returns now form much of the basis of the calculation of New Zealand’s GST, because the GST concept is almost identical to the production-based GDP concept.

Another way of thinking about the role of intermediate consumption in final GDP is to think about costs and benefits. The benefit of production is the final sales price less the costs of producing the goods and services sold. From a national perspective if sales of production remain the same but the costs of production fall, then the net benefits of that production have increased. So, returning to the example of a dairy farmer, if a new technology, for example, allows farmers to use less fertiliser to produce the same amount of milk, then GDP will increase. The reason that it has increased is that while the final sales revenue has remained the same, the level of intermediate consumption has gone down.

The important point here is that GDP is only measuring what GDP is supposed to measure. This might sound like a tautology, but it is an important tautology.  GDP was never designed to be a measure of overall well-being. Indeed, one of the pioneers of the techniques of GDP, the American economist Simon Kuznets, makes this clear in the very first set of national accounts of the United States that he presented to the Department of Commerce in the 1930s.

Professor Woodford makes much of the suggestion that intermediate consumption is a benefit to the economy. This is certainly true in a general sense: the people providing the intermediate goods are paid for doing so and so benefit from the activities. When Professor Woodford says that the GDP figure we included in our report did not take into account the considerable additional economic resources used in producing that output, he is perfectly correct.

He is however incorrect to suggest that the contribution of other sectors of the economy such as manufacturing and other services be allocated to other components of GDP. No output of any sector of the economy that does not represent final consumption by final consumers is included in any aspect of the production-based measure of GDP, nor should it be. Once again, this is what GDP was meant to do. It is designed to measure the net contribution of all the economic activity to all of the economy.

While it may come as a shock to non-economists, much of what is referred to as the “productive sector” of the economy doesn't get included in the final GDP measure. All the outputs from factories, manufacturing plants, offices, and shops that are purchased by other commercial enterprises and then combined to make additional outputs, will never make a contribution to GDP. What matters for GDP purposes is the value of sales to final consumers - individuals, not firms.

For all of the above reasons, 3.09% is the correct value of dairy’s contribution to the New Zealand economy.

Professor Woodford talks about the agribusiness concept and points to research that emphasises that the agricultural sector is really a network of independent activities spanning farm inputs, from production and product processing to marketing. He then says “if you want to measure the value of this interdependent system then we have to look at the total value of output”. The key issue is what do you mean by “total value”. It would appear that what Professor Woodford is saying is that the costs and benefits of producing agricultural output should be summed together to get this measure of value. Here we must respectfully disagree.

As economists, we think that what is important to an economy is to maximise the revenue gained from selling products, while minimising the cost of producing those products, particularly if you are worried about sustainability and the ability of physical and natural capital to continue producing economic well-being now and into the future. This is what the statisticians’ measure of GDP is meant to do. In considerable amounts of political and social discourse both here and overseas, it is common to see a confusion between the value of production and its costs.

The second issue raised by Professor Woodford is the value of exports to the economy. This is often an issue of the technical terms that economists use when thinking about preparing the national accounts getting confused with the everyday meaning of those words.

The value of exports is, of itself, not part of the production-based measure of GDP. Domestic and overseas residents make the same contribution to measured GDP when they purchase goods or services that originate in NZ. We see frequent discussions, some of them in official government publications, that suggest that the value of exports is the true measure of economic performance. Professor Woodford himself says “I also note that it is exports that drive the New Zealand economy”. Again, we must respectfully disagree. There is in fact no single driver of the NZ economy. The economy is made up of thousands and thousands of firms, some big but most small, buying and selling from numerous other traders and then making final sales to individual purchasers. The final sales that are made to consumers of goods and services produced in New Zealand is, as Professor Woodford points out, the result of a complex interplay between numerous intermediate producers trading amongst themselves. What matters when it comes to measured GDP is the final value of sales to final consumers, less the cost of intermediate consumption.

It is simply not the case that exports of dairy products, exports of services like the sales to tourists that are now a significant part of our economy, or any other exports are more important to the NZ economy than any other class of exports. In principle, all make an equal contribution to GDP.

We would stress that our report was not intended to say that the dairy sector, or indeed the wider agricultural sector, made no contribution to the New Zealand economy. That is, of course, a ridiculous proposition and one that we would be the first to dismiss as wildly inaccurate. What we wanted to do was to put the contribution that dairy and dairy processing make to the economy into its correct context, using standard measures of economic activity. What the data that we presented showed is that the combined net value of dairy and dairy processing to the New Zealand economy is about 3.09% of GDP. That is a simple statement of fact. That it may take billions of dollars of intermediate consumption to produce the final products of the sector is not something that should be added on to GDP to get a clear sense of the total contribution of the sector to the economy. This is what Professor Woodford seems to be suggesting.

Likewise, we were motivated to make the point that it is not just the value of exports that matters. What is important is the revenue from exports less the cost of producing those exports. Again, to somehow think that increasing the cost of producing our dairy products or indeed any other export, might lead to an increase in economic well-being is simply untrue.

GDP is an imperfect measure of well-being. Some of the points Professor Woodford has made amply demonstrate that point.

GDP is an accounting measure of the net return to an economy of its production. While GDP does accurately measure the market value of outputs and the market value of inputs, there is much more to life, even economic life, than is included in those two measures. GDP records what consumers pay for the products that they purchase, not the value that they receive from consuming those goods. On the input side, one of the clear deficiencies of GDP as a measure of economic well-being is that it only considers the costs of production that are represented by market-based transactions. The social, environmental and other consequences of production are not included.

We readily concede that the economics profession has not done a good job in explaining the important differences between value and costs. We hope that this note will at least go some way to overcoming this deficiency.

[1]  The two other approaches are the income measure, which adds up all the profits and incomes of all participants in the economy, and the expenditure, which measures value-added. These threes measures, while created using different techniques, all produce the same final result.

This article is NZIER's response to one by Keith Woodford, here.