Appendix E

AUSTRALIAN PRODUCTION COSTS OF SUGAR AS BACKGROUND TO POLICY DEVELOPMENT

By Dr Peter Chudleigh, Agtrans Research

Background

It has long been argued that Australian sugar production costs are among the lowest in sugar producing countries around the world. For example, the 1989 report of the Senate Standing Committee on Industry Science and Technology, in its consideration of assistance for the sugar industry, quoted Warren (1989) where the representative from Landell Mills Commodities asserted that only South Africa and part of Brazil had lower ex-mill production costs than Australia. Recently, the International Sugar Organisation (2002) categorised Australia, along with Zambia, Zimbabwe, Guatemala and the Centre/South of Brazil as "low cost producers". The range for the average cost of production over the period 1995/96 to 1998/99 across these countries was given as 7.58 to 8.18 US cents per pound (c/lb).

World sugar prices are likely to sustain a period of uncertainty where they might remain below average prices experienced in the past few decades. It is therefore recognised that the cost of production of sugar in Australia and its potential to be lowered are key factors in responding to such a low price period.

The reality of this situation has now emerged with Brazil's increased sugar output and improved quality threatening Australian industry profitability via world prices and markets. Given Brazil's expected large increases in production, as of June 14th, futures prices for sugar up until March 2004 (New York No 11) do not exceed 6 US cents per pound. There is some opinion circulating that Brazilian sugar production (at least that in the central/southern region) is economically viable at a price of 6 US cents per pound or even less. In contrast, an average price of 7 US cents per pound at an exchange rate of 0.52 US$/A$ has been suggested as the point where the Australian industry (cane farming and milling combined) ceases to make any return to capital invested. This was based on various average and indicative growing, harvesting and milling costs (Keating et al, 2002). Another benchmark quoted by Australian Sugar Milling Corporation is that 7.6 US cents per pound (at US$ 0.50/A$) is a break-even point for the Australian industry. The definition of cost of production can vary so that care is necessary in interpreting these numbers. Further, such prices are sensitive to A$ exchange rates. Recent strengthening of the A$ would have increased these benchmarks.

There are of course variations in the costs of individual cane farms and mills across Australia, and costs of production vary from season to season. The underlying variation in the cost structure of Australian cane farming and milling is a key component in any assessment of assisted structural adjustment as a response to a period of low prices. This variation needs to be understood to the maximum extent possible by region, by farm size, and by management type within the same farm size/region. This brief review assesses the information available to underpin any conclusions about impact and opportunities open to industry and government for developing any structural adjustment policies in cane farming.

Review of Information Available

Unlike other Australian major primary industries, there is very little published information available on costs of production of cane between regions and for different sized farms. The only authoritative study that has been conducted in the past decade has been that by ABARE (1996). This sugar cane industry survey covered the three years 1993-94 to 1995-96 and was undertaken at the request of the Sugar Industry Review Working Party. The survey covered farmers in Queensland only.

A study by Tessema and Topp (1997) used the ABARE data to show that there were significant differences in returns to cane farming between the four Queensland regions (northern, Burdekin, central and southern). A summary analysis of the average production costs for the four regions over the three years of ABARE data is shown in Table 1.

Table 1: Summary of Production Costs Averaged over Three Years ($/tonne of cane delivered by Queensland farmers 1994-1996)

North Qld / Burdekin / Central Qld / South Qld / All of Queensland
Average
Production
(t c per farm) / 6,111 / 10,242 / 7,301 / 4,496 / 6,598
Average cash costs ($/t) / 21.07 / 25.14 / 23.59 / 24.59 / 23.16
Average depreciation cost ($/t) / 3.16 / 2.38 / 3.03 / 3.79 / 3.06
Average imputed family labour cost ($/t) / 4.65 / 3.27 / 4.22 / 6.91 / 4.59
Average total costs ($/t) / 28.87 / 30.79 / 30.84 / 35.30 / 30.81

(Source: Compiled from ABARE, 1996)

Although based only on three years and a restricted number of farms (195 across the four regions), these averages show that cash costs per tonne of cane produced are generally lower in northern Queensland than those in other regions. This was despite the northern Queensland average farm having only a farm production of around 6,000 tonnes of cane per year, the second smallest average production across the four regions. This result could have been associated with the rain fed (as opposed to irrigated) nature of much of northern Queensland production. It should be noted however, that differences in the sugar content of cane between regions is not accounted for in the above analysis. Lower sugar content of cane in northern Queensland compared with other regions could reduce or offset the region's lower cane production cost.

Family labour costs are higher per tonne of cane produced in southern Queensland where farms generally produced less cane per farm than in the other three regions. Further, the southern region had a higher capital investment per tonne of cane than the other three regions.

The figures in Table 1 include interest paid and any land leasing costs (both included in cash costs). On average these costs represented only about $1 to $1.50 per tonne of cane and there is no other allowance in these figures for any return on total capital employed. A 3 percent return on full equity of a $1.2 million average total investment (average cane farm investment as reported in the ABARE survey) would represent on average over $5 per tonne of cane ($36,000/6598 tonnes), or about a further $4 per tonne of cane ($5.50 less $1.50) over and above the figures in Table 1.

It should be noted that 7 US cents per pound translates approximately into a sugar price of about A$300 per tonne and a farm price of $26-27 per tonne of cane at an exchange rate of 0.52 US$/A$ and an average CCS of 13.5. At an exchange rate of 0.56, 7 US cents per pound is about A$280 per tonne of sugar (see Table 2). These translations are only approximate as there are likely to be differences between the US cents market price and the Australian price that is paid by Queensland Sugar Limited (QSL) to millers. This is because of QSL costs in marketing to fob
(-ve), QSL hedging policies (+ve), and any regional premiums that QSL receive (+ve). Also, these translations to cane prices are sensitive to CCS under the existing cane pricing formula.

Table 2: Conversion of US cents/lb Price to Australian Sugar and Cane Prices (CCS =13.5) with Varying Exchange Rates (conversions do not take into account QSL marketing costs, regional premiums and hedging policies)

Scenario 1 US$/A$ = 0.52 / Scenario 2 US$/A$=0.56 / Scenario 3 US$/A$=0.59
US c/lb / $A /t sugar / $A/t cane / US c/lb / $A /t sugar / $A/t cane / US c/lb / $A /t sugar / $A/t cane
4 / 172 / 15.31 / 4 / 160 / 14.26 / 4 / 152 / 13.57
4.5 / 194 / 17.16 / 4.5 / 180 / 15.97 / 4.5 / 171 / 15.19
5 / 215 / 19.00 / 5 / 200 / 17.68 / 5 / 190 / 16.81
5.5 / 237 / 20.84 / 5.5 / 220 / 19.39 / 5.5 / 209 / 18.44
6 / 259 / 22.68 / 6 / 240 / 21.10 / 6 / 228 / 20.06
6.5 / 280 / 24.52 / 6.5 / 260 / 22.81 / 6.5 / 247 / 21.68
7 / 302 / 26.37 / 7 / 280 / 24.52 / 7 / 266 / 23.31
7.5 / 323 / 28.21 / 7.5 / 300 / 26.23 / 7.5 / 285 / 24.93
8 / 345 / 30.05 / 8 / 320 / 27.95 / 8 / 304 / 26.55
8.5 / 366 / 31.89 / 8.5 / 340 / 29.66 / 8.5 / 323 / 28.18
9 / 388 / 33.73 / 9 / 360 / 31.37 / 9 / 342 / 29.80
9.5 / 409 / 35.58 / 9.5 / 380 / 33.08 / 9.5 / 361 / 31.42
10 / 431 / 37.42 / 10 / 400 / 34.79 / 10 / 380 / 33.05

As $30.81 per tonne of cane is the average cost estimated in the ABARE survey (Table 1), a farm cane price of $30.81 less about $1.50 ($29.30) would cover all costs but would not service any capital except for depreciation. This is equivalent to a price of around A$335 per tonne of sugar and this could be taken as an average reference survival price for cane farming as the industry exists today. This price would cover cash costs, depreciation, and an allowance for family labour, but no capital servicing except for depreciation. Further, it would not cover interest payments for many farms that are now more highly indebted than in the mid 1990s.

If no allowance for depreciation is made (any capital replacement is assumed deferred), the survival reference price would be in the order of $3 per tonne of cane less, or around $300 per tonne of sugar. Obviously the farm business could not be sustained for very long without capital replacement. It should be noted that these reference prices are based on average costs from the ABARE survey.

It should be recognised also that input costs and productivity might have both changed since the ABARE survey in the mid 1990s. If the estimates in Table 1 are to be used as indicative of current costs, it needs to be assumed that any cost increases since 1995 have been offset by productivity improvements. The price estimates assume a direct conversion from the world price for sugar and that paid to millers. The reference points can therefore only be considered indicative.

There is no other recent authoritative industry-wide data available on farm costs than the ABARE data presented. However, it should be noted that CANEGROWERS is currently conducting a farm economics survey that will improve this situation.

Economies of Size

ABARE (1996) concluded that that there was no empirical evidence of economies of size in Australian cane farming. This was based on an analysis of the cost of production data from the ABARE survey for the year 1994/95. A more detailed report by Tessema and Topp (1997) provided a similar view and a useful frequency table of average costs by size of farm (Table 3). These data showed that there was no consistency in cost of production by size of farm, and that considerable numbers of very efficient producers fell in each size category.

Table 3: Distribution of unit cost of production by size of farm

Area / <34 ha / 34-45 ha / 45-65 ha / 66-94 ha / >94 ha / Total
Unit costs ($/tonne) / % farms / %farms / %farms / %farms / %farms / %farms
Less than 23 / 2.7 / 4.9 / 2.6 / 5.0 / 4.9 / 20.1
23 to 26.5 / 3.0 / 4.4 / 3.3 / 5.3 / 4.0 / 20.0
26.6 to 30.0 / 4.5 / 4.9 / 4.9 / 2.3 / 4.0 / 20.6
30.1 to 35.2 / 4.0 / 1.9 / 5.1 / 3.5 / 5.0 / 19.5
>35.2 / 6.0 / 3.6 / 3.9 / 4.0 / 2.6 / 20.1
Total / 20.2 / 19.7 / 19.8 / 20.1 / 20.5 / 100.3

Source (Tessema and Topp, 1997)

A likely explanation suggested in the Tessema and Topp study is that there are mechanisms that smaller farmers can use to avoid capital and labour lumpiness such as strategies of contracting, working off-farm, and also being more flexible, timely, and careful with their own labour inputs to the cane enterprise. The ABARE conclusion of absence of economies of size was used in the Report of the Sugar Industry Review Working Party (Vaughan R B, 1996, p203) to argue that the cane assignment system was not impeding the achievement of size economies in cane farming.

On the other hand, a different position was taken by the Industry Commission (IC) (1996) where it was argued that, based on a small case study, the minimal efficient cane farm in the Burdekin area was 300 hectares (ha) and that the assignment system was impeding the achievement of economies of size. A similar position was argued in a Centre for International Economics (CIE) report (1991) where it was intuitively assumed that there were economies of scale that were not being exploited (CIE, p21). Hanlon et al (2000) suggest there are significant economies of size in cane farming. They quote total cost of production of $18-20 per tonne of cane for very large farms (100,000 tonnes) but up to $35 per tonne of cane for very small farms (2,500 tonnes). As with ABARE, they stress that there are efficient producers in all size categories. The indicative data of Hanlon et al were based on 127 farms that participated in a benchmarking project, rather than being drawn from a random sample. The farms were spread over the Herbert, Burdekin, Bundaberg and Maryborough districts.

The available published conclusions are not consistent with one another with regard to size. However, the farm sizes analysed from the ABARE survey data cover only broad groupings of size with the largest farms grouped into the greater than 94 ha category. If economies of size apply to farms towards the end of, or even outside of the current industry size distribution, any economies of size might not be demonstrated in the analysis.

There is an even more powerful argument for pursuing changes to structures in the industry associated with farm size. The existing cost data assembled are influenced by the past set of industry arrangements and structures that are based on equity and sub-optimal efficiency goals, rather than whole-of-industry efficiency. Changes to these arrangements might well result in economies in farm size becoming more apparent than hitherto. For example, much sugarcane is harvested by contractors whose charges do not often discriminate between blocks and farms that impose different costs through differences in length of rows, proximity to the next block or farm, shape of block etc. Such averaging and the implicit cross-subsidies would be favouring the smaller blocks and farms.

Conclusions from this Analysis

Some conclusions are:

(i)There is some evidence that economies of size in cane farming exist and can be exploited in any market led or assisted structural reform.

(ii)To be effective, any initiatives to rationalise farm size with regard to lowering industry costs should include changes to organisational arrangements such as for harvesting. That is, if policies covering exit and aggregation with regard to farm size are not undertaken in conjunction with other structural reforms, they might be less effective.

(iii)Small farms without off-farm income are likely to be currently struggling to provide living expenses for their families, more so than larger farms, as any margins above essential production costs are likely to be lower.

(iv)Farms in the southern Queensland region are likely to face more severe hardship in a price downturn than in other regions due to their higher costs of production per tonne of cane. This region also has significantly higher total investment per tonne of cane produced and could have more difficulty in servicing any return to capital, whether it be equity or loan capital. This situation is currently exacerbated in parts of the southern region by the lack of irrigation water available.

(v)Although information is currently unavailable on debt levels by farm size, it is likely that farms of all sizes with high debt levels will experience severe financial difficulties; the southern region will be particularly vulnerable to debt servicing problems.

(vi)An average reference survival price is probably around $300-335 per tonne of sugar. This is equivalent to between 7 and 8 US cents per pound at an exchange rate of
0.52 US$/A$, and assuming a direct conversion of the world price to that paid to millers by QSL. Those farmers who can further defer capital replacement, those who have
off-farm income, and those with minimal or no debt would lie towards the lower end of this reference survival price range ($300 per tonne). Those small farms with no off-farm income, and those of any size with significant debt will fall at the upper end of the range ($335 per tonne) and will have a survival price possibly much higher. Also, it should be recognised that such a reference price is an average and such a price will vary for individuals due to other factors apart from those mentioned above, as well as with the seasonal conditions that will influence cane yield and CCS.

(vii)The analysis of ABARE data (Table 3) indicated that 20 percent of farms in the survey had cost of production less then $23 per tonne, while at the other extreme, 20 percent had costs greater than $35 per tonne, with the remaining farms in between. The report of Hanlon et al also indicates that there are large differences in cost of production between cane farmers.

(viii)In general, available authoritative industry-wide information on costs of production by farm size and region is outdated and inadequate. It is understood that this is currently being addressed through a CANEGROWERS farm survey.

Information in Submissions

A number of the submissions received in the current Assessment commented on size and the family farm.

The submissions strongly indicate:

(i)there is widespread recognition that some form of rationalisation of cane farming is required, particularly in relation to farm size;

(ii)the family farm is seen by some as a strength of the industry in terms of its greater flexibility and hence ability to survive downturns;

(iii)while small farms might require off-farm income to be self sustaining in providing a living for a family, some can be efficient in terms of cost of production as presently measured;

(iv)there is general recognition that many farms can not survive the expected fall in incomes, particularly small farms without off-farm income and those farms carrying large debt;

(v)there is some concern that large corporate or cooperatively run farms might be less effective in supporting the existing community fabric; and

(vi)the range of costs of production reported in the submissions is consistent with those in the ABARE survey.

Debt

The Queensland Regional Adjustment Authority (QRAA) debt information provided to CANEGROWERS shows that debt has increased in the past two years from December 1999 to December 2001 by $152 million (m) (from $1,028 m to $1,180 m). The debt when the first survey was undertaken in 1994 was $512 m so that there has been an increase of $668 m over the past seven years.

QRAA data show that there are currently 2,751 borrowers in the industry, similar to the last review when there were 2,784 borrowers. The average debt per borrower is currently $428,000, up from $369,000 in 1999. In addition the rating of debt has deteriorated in the past two years with a significant increase in B1 debt, defined as "borrowers who are experiencing debt servicing difficulties and a deteriorating debt situation but with continuing support from lenders".

Table 4: Cane Farm Business Debt

1994 / Dec 1999 / Dec 2001
Total farm Debt (m$) / 512 / 1,028 / 1,180

Source: QRAA

General Conclusions

  1. There is likely to be serious hardship faced by farmers given the expected low prices of sugar. Farms that will be most vulnerable to any sustained downturn in price will be:
  • smaller farms without off-farm income, particularly those small farms in the southern region; and
  • any sized farm with significant debt exposure.
  1. In order to lower the cost of production it would be helpful for industry to pursue strategies that encourage an increase in farm size. This will enable the industry to take advantage of any size economies that exist within the current organisational structure of the industry.
  1. To be effective, any strategies to rationalise farm size should be associated with changes to existing organisational arrangements that impose added costs on the industry as a whole. If strategies and accompanying policies addressing exit and aggregation are not undertaken in conjunction with other structural reforms, they might be far less effective.
  1. Presumably the definition of the family farm used in the submissions is to do with sustaining a living for the owners of the farm, where the owners live and work on the farm. A
    non-family farm is viewed as a corporate or company farm that is only profit motivated and where the mainstream labour input is hired. While the corporate structure is rare in cane farming, there appears a serious concern that movements in this direction will occur and will be detrimental to the social fabric and local community.
  1. Small farms that are efficient and sustainable should not be "forced out" by industry and government restructuring policies. In other words, the very productive, efficient and hardworking farmers who might or might not have off-farm employment and that are seen by many to provide flexibility and strength to the industry, should be able to continue if they wish. This is provided the real cost of their operation and size is reflected in the economic costs and hence charges they face.
  1. There is no evidence to support an optimal or minimal farm size that should be targeted.
  1. There is evidence to show that the difference in costs of production between farmers varies widely. It is concluded that there are likely to be significant potential savings on an industry basis through management improvements.
  1. Industry information on costs of production by farm size and type is outdated and inadequate and steps should be taken to remedy this situation. The current CANEGROWERS survey should improve the situation, but economic information should be assembled in an ongoing and consistent framework.

References