I have just come across a paper I wrote in the 1990s that has continued to subsist on the net: The Construction and Use of Coincident Indicators Based on the Data Contained in Business Surveys. It remains alive since it was a paper I presented to the OECD which has meant that it still has a lingering life out there in the net.

It’s not as if I don’t remember the paper, which I do. The OECD even took action based on it, and began to pay more attention to business surveys along with the usual consumer confidence surveys which I thought then (and to the extent I still think about such things still do) as utterly useless in trying to stay on top of the business cycle. The article begins:

Knowing how well an economy is performing is a far more difficult task than it often appears to be. There are so many aspects to economic conditions that it can be extremely difficult to gauge with any sort of precision our actual position at any given moment in time.

There is also the problem that for most of the major activity measures, such as the level of national output or the rate of investment, official figures are at the minimum three months out of date on the day they are published, and even then, if we are to identify an actual trend, the data may well be half a year old before one can infer that the direction of the economy has changed.The National Accounts are the most comprehensive measure of economic activity available within any economy. They are also the most watched measure because they bring together into a single framework all the different aspects that make up the total level of economic activity.

Yet aside from the delay in having such data published there is also the “noise” that surrounds the figures each time they are produced. Because the National Accounts are a complex array of different features of the economy, the underlying direction of economic activity is often obscured. That this is the case is widely understood which is why a slowdown in a single quarter is never taken as in itself evidence that the economy is indeed slowing. The random factors which affect the measurement of not just the total level of activity but also each of the individual components make it impossible to draw any hard and fast conclusions about economic conditions from quarterly shifts in national accounting estimates.

The rest of the article presents my approach to keeping up with the direction of an economy. This is the conclusion to the article:

The kinds of series that can be developed from business surveys allows such surveys to be used in a more systematic way than at present. They provide information that is timely and because of the form of construction provide an indication of the strength of the current trend in economic activity.

The accuracy of the barometer over the past not only provides strong evidence that it will continue to provide genuine assistance in measuring the strength of the economy, but it also demonstrates the value and accuracy of the business surveys which lie beneath. These surveys provide comprehensive data on the condition of the economy and do so far earlier than any other source.

Although official statistics provide a series of partial indicators on the level of various aspects of the economy, they do not provide a full range of indicators on conditions overall. It is why business surveys are so important to the stability of an economy because they provide virtually immediate information on current conditions available from no other source. But they provide information only to those who are attuned to the messages they send. One obviously cannot react to every fluctuation in business surveys as they, in company with official statistics, have a random component which will suggest strength where there is none, or weakness when the economy is strong. However, over time, these are a critically important measure of the health of an economy that the barometer allows for presentation in a more systematic way.

And because the barometer is focused on the private sector, and is not obscured by data from non-market activities, it provides a more concentrated view of current economic conditions. The strength of this barometer lies in the basis of its collection. The views in such surveys are those of the chief executives of businesses. These are the decision makers across the economy and it is they who determine the level of output, investment and employment.

These surveys tap into a different form of data from those found in official statistics since they seek from those who make these decisions an indication of the decisions they are in the process of making. This is information of the most crucial kind and the results of such surveys deserve to be listened to closely. The Business Barometer systematises the views of those who own, operate and manage firms. They therefore provide the data necessary for policy making in a form that is easy to understand and which can be incorporated into the official family of major statistics used for assessing economic trends.

Alas, the problem with the Business Barometer was that it was too accurate. It annoyed Treasury and the Reserve Bank at the time since there we were, this miserable business association, able to accurately foretell turning points in the cycle. I will just leave this out there just in case someone might see this and think they might give it a go.