Appendix 1: Methods for evaluating the National benefit from irrigation

1 Introduction

This section describes alternative methods considered that could be used to calculate GDP and identifies the reasons for choosing the adjusted gross margin method. The adjustments made to the gross margins are also outlined.

2 Valuation of Water[1]

The System of Environmental and Economic Accounts (SEEA) Manual[2] lists three ways that water can be valued:

(i) Using the value of short-term or perpetual water right rentals.

(ii) Comparing farms without irrigation to identical farms with irrigation.

(iii) Using a cost approach: i.e., estimate the value based on the cost of supplying water.

The first method is impractical for New Zealand due to the absence of a market for water rights.  The SEEA notes that the third approach is "the least satisfactory from a theoretical point of view". Method two is therefore the most satisfactory approach under New Zealand’s institutional framework. The SEEA notes practical problems with this methodology, particularly whether farms without irrigation can be validly compared to farms using irrigation, especially when land use changes as a result. However, provided equivalent farms are compared, the increase in farm value added can still be attributed to irrigation.

3 Calculating contribution of irrigation to GDP

GDP is a measure of the performance of the New Zealand economy. It is an aggregate measure of the production of goods and services within New Zealand. Many goods and services provided by one producer are purchased by another for use in subsequent production (intermediate consumption) and are therefore deducted from gross output to avoid double counting. For individual producers gross output less intermediate consumption measures their value added and represents that producer's contribution to GDP. For industries, value added equals the value of gross output of each industry less the cost of goods and services used by it in production[3],[4].

In this analysis, two methods were considered for evaluating the contribution of present and future levels of irrigation to GDP. Input-output analysis was one, and the other was a direct assessment of changes in gross output and intermediate consumption using appropriately adjusted gross margins.

3.1 Input-output analysis

An input-output (I-O) table is a model of the economy at a particular point in time. It illustrates the interactions and dependencies between industries. It can be used for economic analyses including[5]:

1. Identifying and measuring the composition and level of economic activity.
2. Understanding the inter-relationships between industries.
3. Studying the effects of changes in supply and demand throughout the economy including total economic activity (output), household income, value added and employment.
4. Analysing the flow of goods and services between industries and final users.
5. Providing the basis for the calculation of Gross Domestic Product (GDP).

Consideration was given to using I-O analysis as a tool for estimating the impact of irrigation on GDP, employment and total output from a region and the nation. I-O analysis has been used to evaluate the regional benefits of irrigation. For example, the Central Plains Water Enhancement: Economic and Social Impact of Proposed Irrigation Schemes report[6] used I-O to analyse the impact of irrigation in Canterbury in terms of output, employment and value added. Specific multipliers were estimated on the basis of irrigated farm model budgets.

However, the limitations of input-output analysis are well documented.

1. I-O assumes that marginal changes are the same as average ones e.g., assumes that there are no price effects as a result of changed output from a sector. This is clearly inadequate in the case of the large-scale changes in agricultural output generated by current and possibly future expected levels of irrigation, at the national level.
2. I-O assumes that the supply of every input is perfectly elastic i.e., there is sufficient surplus capacity in the economy to deal with any increase in output e.g. processing capacity. This is not the case for many agricultural products, for example dairy factories would need to be expanded if output increased substantially.
3. I-O assumes linear production function technology – it doesn’t allow for substitution between factors of production (even if prices change), nor for economies of scale.
4. I-O assumes fixed technology and constant sector purchasing patterns.

The advantage of using I-O analysis in the current study would have been that an estimate of the flow-on effects through the economy of the increase in farm gate value of output and value-added could have been calculated using multipliers. Similarly, increased employment beyond the farmgate could have been estimated. However, robust estimates of the appropriate multipliers are difficult to derive, requiring the development of irrigated and non-irrigated model farm budgets across the range of farming types (and in some cases, regions) covered in the study. This report calculates farm gate values only. An assessment of the flow-on effects of community irrigation schemes in Canterbury may be found in Ford (2002).

3.2 Adjusted Gross Margin method

A gross margin is the total revenue associated with a particular production (income) less the costs that clearly vary in direct proportion to the level of production - the direct or variable costs associated with the enterprise. Gross margins are an accepted tool commonly used in the evaluation of farming enterprises. They have been used for the evaluation of the costs and benefits of irrigation[7] in cost benefit analysis. Assessing the change to the gross margin per unit area as a result of irrigation and then scaling this appropriately by the total affected area provides an initial estimate of the GDP change (at the farmgate) likely to occur as a result of irrigation.

This method of adjusted gross margin analysis accords with the SEEA recommendations (section 2) and provides a “best estimate” of the change in GDP generated by irrigation at the farm gate[8]. However, it should be noted that a large number of estimates and assumptions were required to estimate the impact on GDP, and the results should be interpreted with caution. In addition, the increased output from irrigated farms will have different flow-on effects in the wider economy, so the total impact on GDP is likely to be higher than the farm gate impact.

The Gross Margins were adjusted to take account of differences in overheads between land uses, and also for the treatment of wages and salaries to better approximate the way GDP is calculated in the National Accounts.

Overheads

In a standard gross margin calculation, overheads[9] (which typically do not vary significantly as production levels fluctuate) are excluded from the analysis. The underlying assumption is that overheads are similar between the enterprises being compared. However, for this analysis of irrigation GDP contribution, assuming unchanged overheads may result in significant over estimation of the net contribution. The different land uses assumed in the ‘with’ and ‘without’ irrigation cases usually have completely different fixed cost structures. Failing to recognise this fact, and adjust overheads accordingly between the with/without cases is therefore likely to produce an upwardly biased GDP estimate.

Therefore in this analysis the value of irrigation, as assessed by the change in gross margin analysis, has been adjusted to reflect expected changes to overheads arising from having the land irrigated. This adjustment has been based on known current overheads and typical fixed costs per enterprise[10] in areas currently irrigated.

Salaries and wages

A potentially even more serious source of error arises from the way “salaries and wages" are treated in GM calculations compared to their treatment in a GDP calculation. In a GM these charges vary in their treatment – sometimes they are treated as overheads and sometimes as direct costs of the enterprise being analysed.

To be consistent with how farmgate GDP is calculated in the National Accounts, a distinction has been made between those wages and salaries within the enterprise that would be part of Intermediate Consumption, and those salaries and wages that would remain within the farmgate GDP calculation.

In converting the GMs into a GDP the following rules have been applied.  For dairying, arable farming or pastoral farming and process vegetables all labour has been assumed to be part of the permanent farm staff and wages/salaries paid to these have been added back into the relevant GM's. This adjustment has the effect of raising the farmgate value of GDP of these activities.  For apples, kiwifruit and other horticultural crops it is more the norm to employ contractors to carry out operations and for these crops the unadjusted GM has been allowed to stand and GDP calculated as though all wages were part of intermediate consumption.  That in turn may well mean that the farmgate GDP impact of increased irrigation of these crops is somewhat understated in the body of the report.

4. The Impact of increased output on price

Gross margin calculations also generally assume that a change in output has no effect on prices. While for small-scale changes at the individual farm level this may well approximate the truth, the large-scale land use changes generated by irrigation on the national scale are believed to be sufficient to have some measurable effect on output prices. An assessment has been made of the likely magnitude of the price changes using the Lincoln Trade and Environment model as the primary analytical tool, as described in sections 4.7 and 5.4.

The Lincoln Trade and Environment Model (LTEM) is a partial equilibrium (PE) model, used to quantify the price, supply, demand and net trade effects of various policy and non-policy induced shocks. The LTEM is an agricultural multi-country, multi-commodity trade model. There are 18 countries and 19 agricultural commodities included in the model.  The model is used to derive the medium to long-term (till 2010) impact in a comparative static fashion, basing the beginning date to 1997. A fuller description of the model may be found in Cagatay and Saunders (2003)[11].

As noted in the text, the products covered by the LTEM contribute 45% of the total contribution of present day irrigation to GDP. Other products of irrigated agriculture that are not modelled in the LTEM are grain and seeds, vegetables, flowers, other fruit crops, wine and deer. The prices of most of these products tend to be sensitive to changes in output volumes, particularly those supplying the local market. Intuitively it would be expected that the increases in production generated by irrigation would result in a fall in price of these products. However, the impact of a “without irrigation” scenario implies very significant changes in the nature of these sectors, the impacts of which on price are not straightforward to determine. In some cases the crop would not be produced at all, in which case discussions on “what the price would be to New Zealand’s producers without irrigation” are unnecessary.  The likely impacts by product type are;

• Grain and seed crops: larger areas of grain crops would be produced under dryland conditions in the “without irrigation” scenario, and depending on the impact on production of natural rainfall each year, production and therefore prices would fluctuate considerably from year to year. Import parity prices apply to grain over time, although it is most likely that grower prices would fall below those assumed in this analysis due to the uncertainty of production and therefore risk to grain purchasers. This would have the effect of increasing the value of irrigation to GDP over that calculated in the body of the report. Furthermore, a significant seed industry would not exist in New Zealand without irrigation.
• Process vegetables: It is unlikely that this industry would exist on any scale in the absence of irrigation.  Processors require an assurance of a reliable supply of raw vegetables before investing in processing plant.
• Fresh vegetables and flowers: It seems likely that this industry would be very much smaller in the absence of irrigation, focusing on the local market. The industry would be located in those areas with the most reliable rainfall.  Production levels and quality, and therefore prices would fluctuate violently from year to year, depending on the pattern and quantity of natural rainfall. Without irrigation, quality and quantity would not be sufficiently reliable to support an export trade (currently substantial for some vegetable crops e.g., onions, squash). Export markets usually yield better prices to the grower than the local market. The overall impact of irrigation on price is therefore impossible to determine and may well be neutral or even positive overall.
• Orchard crops: As with fresh vegetables, it seems likely that this sector would be focused on the local market in the absence of irrigation. Production levels and quality, and therefore price would tend to be more variable without irrigation. Currently the existence of an export market (made possible by the reliability of output and quality generated by irrigation) prevents oversupply on the local market in most years, minimising the impact on price of the increased output generated by irrigation. In addition, in most years, the export price exceeds the local market price to growers. As with vegetables, overall the impact on price of irrigation may well be neutral or positive.
• Wine:  It is difficult to establish grapevines in the high quality wine producing areas (which are also the driest areas) without irrigation. Output and quality would therefore both be lower in the absence of irrigation.  Lower output does not necessarily imply a higher price since New Zealand is a very minor player in world wine trade. Overhead irrigation for frost control is also important in maintaining export production levels of this crop. The overall impact of irrigation on price is therefore ambiguous, and may well be positive.
• Deer: The change in output of deer products as a result of irrigation is very small and unlikely to have a significant impact on price.

The effect of future expansion of irrigation on prices of non-LTEM crops is even more difficult to determine than the effect of present day irrigation. For crops that are largely dependent on the local fresh market and for which there is little opportunity to develop large-scale export markets (e.g., green leafy vegetables and potatoes) increases in production tend to have a dramatic effect on price. For crops which are produced under contract (e.g., process crops, arable crops) the number of contracts being offered limits the output of these crops. For process vegetables, export markets (and therefore contracts) are growing, but are dependent on exchange rate factors and market development and processing capacity.  Some non-LTEM crops are exported (e.g., onions, squash, flowers, deer products, avocados, citrus and wine) and increasing the New Zealand output is likely to require investment in market development, and this could have a measurable impact on price. However, a recent review of the historical trend in New Zealand onion prices at FOB could find no significant relationship between the price and New Zealand production of onions (pers comm Irene Parminter, MAF Policy). World production and trade overwhelms the effect of any increases in New Zealand output, for example, in wine production. The price of some products will change but this change is unrelated to New Zealand’s output or irrigation development. In addition, exchange rate factors are a very important factor in the price received.

A complicating factor in assessing the impact of future irrigation-driven increases in output on price is that growers of annual crops are very flexible in the combinations of crops that they choose to grow. If, for example, potatoes are in over supply, growers would switch to another crop which proves more profitable. The crop combinations and gross margins used in the analysis are therefore only indicative of a range of possible crops with similar outcomes. The price impacts of annual crop output increases are therefore likely to be relatively muted compared with increases in perennial crop output, since the latter are less flexible.

A further complicating factor is the increase in flexibility (in terms of crop selection) of the grower/farmer, the degree of management control able to be exerted, and the reliability and quality of output under an irrigated regime. The grower/farmer is therefore able to choose the most profitable product to produce, and to increase the value of the product e.g., by producing at a time of the year when price is highest, or by increasing the quality of the product (for example, through improved fruit size). This upside potential has not by and large been included in the analysis.

In the light of all these considerations, the impact of a 10% fall in price of the non-LTEM crops was modelled when assessing the impact of future irrigation driven increases in output.

4 Measurement of the National benefit from irrigation

The impact of irrigation on GDP is greater than the national benefit from irrigation. GDP represents the return to producers’ labour and capital (including capital tied up in land). The increase in output (and the change of composition of the output) arising from irrigation is dependent on capital investment in irrigation infrastructure on and off the farm, and in on-farm improvements. It is also likely to demand increased labour and managerial input from the owner/operator. The gross margin analysis above captured the off-farm investment in irrigation infrastructure (through the cost of irrigation in the gross margins) but not the on-farm investments, nor the increase in managerial input.

To effectively incorporate these considerations, a cost benefit framework is required. The extra data required for cost benefit analysis is not provided in this progress report. Further work is required to gather and incorporate this information with the gross margin information from this phase.