1993 SNA Update Information - Consistency Issues
The market /non-market distinction
|The SNA makes a distinction between three types of production; market, production for own final use and non-market. This division into three followed long and complicated discussion during the revision process before the 1993 publication about how to deal with production which was marketable but not marketed. The major examples are subsistence agriculture, used for household final consumption and own account construction by enterprises. The name “other non-market” was chosen reluctantly because no better name was then apparent. Unfortunately, this is often seen to imply that this third sort of production is also some sort of non-market production and often text is written as if there are still only two choices, between market and non-market.|
In fact it is possible to enumerate four different types of production.
Type A is of the sort which is indisputably market; the production of goods and services for sale at prices which are “economically significant”. That is the prices are set by the producers to cover all costs and make a profit and consumers have complete freedom about whether to purchase the products and the quantity to be purchased. In SNA terms, the prices are significant for supply and demand and they satisfy the terms of the elaboration given in the delineation paper.
Type B production is production for use by the producer. This may be subsistence agriculture or own account construction, as noted above. The characteristic of this production is that the products are suitable for sale on the market and, in principle, could command an economically significant price. Some construction projects are so specific that it is impossible to say what the corresponding “market price” would be but it is clear that a producer will only undertake the work himself it he believes that there are sound commercial reasons for undertaking this activity in house rather than out-sourcing it. In short, therefore, the criterion that the price is significant for supply is met. If the cost of own account production were higher than a certain level, producers would refrain from the production. The “certain level” is determined by their own demand for the product and thus meets the second criterion for economically significant prices.
Type C production is supplied to individual units other than the producer but at prices which are too low to satisfy the economically significant criterion. It is production by public entities in receipt of sufficient funds from government to make good the shortfall of receipts from sales over costs that they can not be regarded as independent of government.
Type D production is supplied collectively to all units in the economy by units of general government. Although it is stated to be possible for all industrial activity to be carried out on either a market or non-market basis, in fact it is very difficult to imagine provision of public administration and defence being carried out on a market basis. It is possible for government to subcontract the running of a prison service and paying the operator of the prison a market rate, but this is not different in principle to government acquiring stationary and vehicles at market prices. The service supplied by government rather than to government is provided without charge to the public at large.
There are two discriminants which determine these four types of production. One is whether the production is actually marketed or not. The other is whether the production is supplied at economically significant prices.
Type A production is marketed and has economically significant prices
Type B production is not marketed but has economically significant prices
Type C production is marketed but does not have economically significant prices
Type D production is not marketed and does not have economically significant prices.
Looking at this demarcation, it is clear that the discussion leading up to the 1993 revision on whether the production was marketed or marketable was insufficiently clear since it did not make adequate distinction between type B and type C production though this distinction is fundamental to the sectoral allocation of units. From this enlarged distinction some other consequences follow.
For type B production, it is not clear that it needs to be restricted to production used for final demand. The SNA refers in para 5.11 to the production of milk and subsequent conversion to butter or cheese, for example. The discussion on ancillary production and on the output of holding companies, SPVs and multi-territory enterprises suggests that production for use as intermediate consumption by another unit in the same group of enterprises, or within a multi-establishment enterprise, can be accommodated in this type of production without difficulty.
Type C and type D production together constitute production by units allocated to general government. This reveals that the present expression “non-market” does not bear on whether the product is marketed or marketable, or on how much is marketed, but simply on whether it is supplied at an economically significant price. Further, and rather neatly, type C production corresponds to individual consumption and there might be corresponding type A production. Type D production corresponds to collective consumption and there is no similar type A production.
In the paper on the informal sector, there is discussion about the market criterion that all or most production needs to be marketed. This criterion for market production can be relaxed; it is only that there are economically significant prices which is important. For the informal production, al that is needed is that the prices are economically significant and at least some is actually marketed. This does not then need to introduce a different criterion for a “market producer” in the household sector.
|The AEG agreed further elaboration of the distinction between market/non-market activities might be useful and invites comments on the proposals put forward.|
Comments are welcome preferably by the end of July but no later than mid-September 2006.
|Number of comments: 16|
| Date posted||Source||Comment|
| 04/10/2006||Jan Heller||I agree with Anne's conclusion|
| 04/10/2006||Irena Tvarijonaviciute||Sorry for the late contribution, but looking at the discussion concerning Market/non-market
I do not agree with a treatment of own account work as a market production. |
I expressed this opinion already in Frankfurt discussing about labour units treatment.
I really support Robin's explanation below and Ole opinion. I do not agree to treat a household
which produces for own final use as a market producer.
| 04/10/2006||Robin Lynch||Disagree.
My understanding of an economist definition of a market is defined by the two characteristics as follows: |
In an active effective market place
1. suppliers have freedom to set a price on goods and services available to all customers
2. customers have freedom to choose amongst all suppliers
Non-market activity takes place when either or both of these conditions fail.
Own-account production cannot be defined as market production, as the product is not sold on a market: there is no observable price, and there is no choice of supplier
Of course if the internal unit supplying the own account work also operated in the market place, and the purchasing unit had genuine freedom to go to the market or choose the internal supplier, then the own account internal supplier could be said to be operating in the market, and a price for the product could be observed in the internal unit's market operations. So there might be special cases where the own account production could be called a market transaction.
But there would be special cases, and the general case is that own account production would be non-market.
The current use of the term "economically significant price" as an aid to determine when transactions are market or non-market is not helpful, and it would be better to revert to the fundamental characteristics of what makes a market as set out above.
So I find the current SNA 1993 definitions very acceptable, and any move to call own account production market as opposed to non-market against fundamentals of economic theory.
Some comments on the supporting paper "Market and Non-market"
1. The two determining characteristics for a market transaction agreed for the 1993 SNA seem strange. Why not stick to the economics definitions?
Supplier free to set a price, customer free to choose between supplier, all taking place in a market where transactions can be observed.
This puts collective services as non-market, own account production as non-market.
2. If the answer to the first question (is the product delivered to another unit?) is yes, we have "production for own final use" How can it be production for own final use? Presumably this is a typo, and for yes we should read "no",. (Or I am really confused!?)
3. How can we treat collective production as provided to other units if the accounts show them as being consumed by the provider by convention. It is strange to have a definition at odds with the framework of the accounts, which shows collective production as being consumed by the producer, government.
4. I disagree with the section on the rationale for treating own-account production as market. How can households that don't operate in the market be market units?
Of course there are always boundary issues about those annoying units that operate partially in the market, but if a unit only produces output on own-account, then it does not operate in the market and so - why should we want to call it a market unit?
Comments on the summary of changes proposed
1. "Production not immediately delivered to another unit is 'own-account production'. So production intended for sale but held in stock as finished goods is own-account production?
2. . I like the term marketed instead of market this will help to emphasise that it is how delivery takes place that determines market or non-market, and not some intrinsic characteristic of the good, service or unit involved.
3. The ESP gets in the way of more fundamental characteristics of a market. In a free market where customers have a choice of suppliers, supply at non-ESP's would wreck the market - black markets would be introduced (cf war-time efforts to ration).
4. To me this reads "market production is defined to cover non-market activities such as own account production. I just don't get this.
5. No comment
| 02/10/2006||Ole Berner||Disagree: I support the distinction between the three categories of production and the revised terminologi (production for sale, production for own-account and non-market). I disagree with the proposal to classify both production for sale and own-account production as market production.|
The argument for classifying own-account production as market production is that it could be sold on the market and that it (indirectly) has an ESP for the producer. Although completely correct I do not agree with the line of argumentation because I think that using these arguments could lead to the conclusion that a large part of government production of individual services (Health, education etc.) also should be clasified as market production. I would propose that in relation to production we just have the three categories.
In relation to classification of units we could then establish a rule saying "Households (or units(?)) producing entirely or mainly on own account are classified as market producers"
| 02/10/2006||Brent Moulton||I agree with the recommendations.
| 30/09/2006||Peter van de Ven||Agree.|
| 29/09/2006||Bank of Korea||We agree to all the recommendations and hope that more research will be undertaken. |
| 29/09/2006||A.C. Kulshreshta||I agree with the
ISWGNA proposal and formulation made by Anne|
| 29/09/2006||Johan Prinsloo||I broadly support the conclusion made by the ISWGNA, however, the meaning of the term “for sale“ should be clear incorporated in the text as mentioned in the conclusion.|
| 28/09/2006||Reimund Mink|| I agree with the ISWGNA proposal on the market/non-market distinction (the last paragraph of the ISWGNA conclusion should read: at non-economically significant prices).|
| 27/09/2006||Mariam Cover||I agree with what is said in the conclusion posted on the website. I find a bit confusing the difference in terminology used in Anne’s paper posted on the web and that used in the conclusion, in the paper, production delivered to another unit at ESP is described as “marketed production”, while in the conclusion it is referred as “production for sale”, I find this second term easier to translate and to understand.|
| 26/09/2006||Heidi Arboleda||I agree with the conclusion made by Anne|
| 20/09/2006||Roberto Luis Olinto Ramos||The clarification of the differents types of production is a need in the SNA. I support the Anne Harrison paper five final recommendations. |
My only concern is how to translate "marketed" to portuguese.
| 15/09/2006||Johann Prinsloo||The revised suggestions on this issue are very practical and I fully support the proposed changes as summarised in points 1 to 5 of Anne Harrison’s revised version.|
| 12/09/2006||Heidi Arboleda||The need to distinguish market and non-market is useful in determining the value of gross output of a producer. The former is computed based on market prices or prices which are economically significant and the latter based on the cost of production. |
I agree with the categorization into Type A-D to clarify the distinction between marketed products and product at economically significant prices. However, the distinction could be made simpler for the purpose of estimating the value of gross output. A and B will be estimated based on the prices in the market of the commodities or if there are none, at prices, which will cover the cost plus operating surplus. C and D will be estimated based on the actual cost without operating surplus/profit. In the case of C, the disposition of the output will be shared by the purchasers (households) and the producers (government or NPISH) and will be reflected in their respective final consumption expenditure. In short the distinction for purposes of estimation is really on whether the product is produced at economically significant prices and not whether the goods are marketed or non-marketed.
I guess there would be no need to change the term but there is a need to change the meaning of the term market and non-market. Market output will therefore refer to those produced and valued at prices at the market or at prices that covers the cost plus operating surplus and non market products are those that are valued based on cost of production without any imputation of operating surplus/entrepreneurial income/profit (there will be a need to clarify this)
| 28/07/2006||Anne Harrison|| MarketNonMarket; |
|It is concluded to continue the separation of production into three categories. |
The first, previously called “market” production, continues to cover production of goods and services intended for delivery to other units at economically significant prices. No change in coverage is agreed on. It is concluded to change the terminology from “market production” to “production for sale”, the term “for sale” being recognised as short-hand for “delivery to other units at economically significant prices”.
The second category relates to goods and services intended for use by the producer. In general, goods produced and then used within the same unit as intermediate consumption are recorded as neither intermediate consumption nor output. However, if research and development, for instance, is not treated as capital formation, and the activity cannot be separated into a different establishment, then it will be shown as a secondary product consumed by the same unit. It is therefore concluded to change the coverage of this item to allow these exceptional types of production for own intermediate consumption. It is concluded the terminology changes from “production for own final use” to “production for own use”.
The first two categories together, that is production for sale and production for own use together form market production.
The third category of output, whereby goods and services are delivered to other units and non-economically significant prices, previously described as “other non-market production” is to be re-titled “non-market production”. No change of coverage is agreed on.