• Rural NZ
• Adverse Events
• People issues
• Profitability
  • Compliance costs
  • e-Commerce
  • Employment
  • New industries
  • Performance
  • Producer Boards
  • Stocktake
  • Structural Change
  • Trends
• Research
• Rural Proofing
• SLM Hill Country Erosion Programme
• Community Irrigation Fund
• Stats & forecasts
• Sustainability
• Legislation
• Media Centre
• Publications
• What's New

---

• Viability and Variance
• Indicators of Business Viability
• Income Generation and the Cost of Production
• Cost of Business
• Household
• Factors Contributing to Variation in Economic Performance
• Physical Characteristics of Farms - Quartiles grouped by EFS
• Physical Characteristics of Farms - Quartiles grouped by Rate of Return
• Physical Characteristics of Farms per $Million Farming Capital - Quartiles grouped by Rate of Return
• Comparison of the Analyses
• Equity in the Farm Business and Home
• Equity in the Farm Business and Home - Quartiles grouped by EFS
• Equity in the Farm Business and Home - Quartiles grouped by Rate of Return
• Farm Business Operation and Efficiency
• Comparison of the Analyses
• Sources and Application of Funds - Quartiles Grouped by EFS
• Sources and Application of Funds per $Million Farming Capital - Quartiles Grouped by Rate of Return
• Sources of Funds - Comparison of Analyses
• Application of Funds
• Change in Equity - Quartiles Grouped by EFS
• Change in Equity - Comparison of Analyses

---

Appendix III: Economic Dimensions

Most New Zealand farmers and commentators rate adequacy of financial profitability as the key driver of environmental and social sustainability. (Underwood 1999;4;1). This component of the report sets out to examine the profitability of North Island hill country farming and establish those factors and trends associated with its viability. Disposable Profit is used as the primary indicator. In addition, Rate of Return (ROR) is used to identify those factors associated with the economic efficiency of North Island hill farms. The overall aim is to establish the sustainability of North Island hill country farming.

A description of the data used in the analysis is shown in Appendices I and II, and definitions of the factors used are provided in the Glossary (Appendix IV). Data were obtained for the 21 years from 1976-77 to 1996-97. This allowed comparison of the pre 1984-85 period with the post restructuring phase. Where real dollar values are used, data were adjusted to 1997 dollars using the Consumer Price Index. This was chosen because most of the data evaluated in this way (e.g. Disposable Profit) involved amounts remaining after farming costs had been subtracted.

Viability and Variance

Since 1976 there has been significant reduction in the level of real Disposable Profit on Class 3 and Class 4 farms (Figure 1). In 16 out of the 21 years most farms were below the mean. In the other years most farms had low Disposable Profit. In effect, when times were average or good, a small proportion of farm businesses did well, but in the tough years a small proportion incurred large losses.

Figure 1: Real Disposable Profit ($ 1997) - Mean and Median by Year

In a presentation to the Wairarapa-Tararua Drought Recovery Programme's Debt Seminar in Dannevirke in 1998, Pita Alexander, stated that "Since 1987 there has been a polarising of the top and bottom farm sections re their financial results - the gulf has been getting wider but the pace of this change has moved up several notches in the year just ended. The top 30% group's results are quite sound to very sound but the bottom 30% group is unsound to very unsound. This trend is just a result, I think, of non-subsidised, deregulated agriculture but it has a lot of implications for a lot of people, particularly if the pace of the change quickens." He added that while the intermediate group made up 40% of his practice in 1998, just ten years earlier in 1987, it had made up 80%, with 10% of farms being financially sound and 10% unsound.

If this view were valid for all New Zealand farms, it would be reasonable to argue, that inspite of the large number of financially unsound sheep and beef farmers exiting their farms in the last 11 years, the proportion and number of unsound and at-risk farms is now much greater than before. Moreover, this trend is probably increasing but because the number and proportion of farmers doing well is also greater than before, reported averages have hidden this fact.

To test this hypothesis, Willis (1998) made a preliminary analysis using the 1997-98 budgets prepared by Agriculture New Zealand for the MAF Farm Monitoring Programme. The 95 farms representing seven sheep and beef farm models in the southern half of the North Island were sorted by Disposable Profit into four quartiles and quartile averages were derived (Table 1). The range identified is larger than expected.

Table 1: 1997-98 Farm Monitoring - Average Disposable Profit by

The standard deviation of the 1997-98 Disposable Profit for the Farm Monitoring farms is $44,664. For the 1998-99 budget period it was forecast at $56,553. In comparison, the standard deviations of Disposable Profit (in real $1997) for MWESNZ farms for the last six years vary little (Table 2).

Table 2: MWESNZ Farms - Standard Deviation in Disposable Profit (real $1997)

In effect, and contrary to Alexander's anecdotal evidence, there is no statistical basis to claim a significant increase in the variability of North Island hill country farmers' disposable profits in recent years; quite the opposite. From 1976 to 1997, the trend in farmers' disposable profits has been downward, and since 1984-85 less variable than before (Figure 2). Whereas the average standard deviation from 1976-77 to 1984-85 was $66,854, it averaged only $49,229 over the period from 1985-86 to 1996-97.

Figure 2: Real Disposable Profit ($ 1997) - Mean and Standard Deviation by Year

Comparison of the Disposable Profit for each of the four quartiles in the 21 year period reveals considerable annual variations. These are particularly pronounced in the top and bottom quartiles. However, while Disposable Profit for the top quartile has declined overall, the bottom quartile has continued to perform within a relatively narrow band (Figure 3). For the full period from 1976 to 1997, the annual rate of decline for top quartile farms was $3,181, for second $1,646, for third $1,091, and for bottom $253 (Figure 4).

Figure 3: Disposable Profit ($ 1997) - Means of Quartiles by Year

Figure 4: Disposable Profit ($ 1997) - Differences between Quartiles over 21 Year Period

Trends were also calculated for shorter time spans within the total 21 year period. For the second, third and bottom quartiles, regression lines differed little from those for the whole (Figure 5). The Disposable Profits of top quartile farms fell sharply from 1984-85 to 1990-91, but were relatively steady before and since then. It would seem that the restructuring, begun in 1984, had the greatest impact on the disposable profits of top quartile farmers. At the same time more farmers exited the industry in the 1984-5-1990-91 period than before and after that time. Presumably most of these were from the bottom quartile.

Figure 5: Disposable Profit ($ 1997) - Regression for each Quartile by Period

In this analysis each year's average Disposable Profit is defined simply as the mean of the quartile for that year. Obviously the circumstances of individual farmers can change widely from one year to the next, placing them in different quartiles. This variation in annual performance, and quartile placement, was explored using the budgeted Disposable Profit of MAF Farm Monitoring hill country farms in southern North Island for the three years from 1997-98 to 1999-00. On this basis, over 40% of the farms in the bottom quartile in any one year remained there in the following year. In effect up to 10% of all farmers were making repeated losses from farming of $20,000 or more per year.

Indicators of Business Viability

A range of farm characteristics - physical features, income, production costs, financial changes and household drawings and other factors - provide a basis to measure the viability of the farm business.

Physical Factors

The physical characteristics of North Island Hill Country farms have changed subtly over the 21-year period.

Average farm business size, as indicated by stock units, has increased by only 5-6%, a trend consistent across all quartiles (Figure 6). This represents an annual increase of 0.3%, less than one third of that identified by Neild and Butcher (1999) in their analysis of trends for all sheep and beef farms over a 25 year period.

Figure 6: Total Stock Units - Means of Quartiles by Year

Farms in the top quartile have consistently managed from 20% to 48 % more stock than the overall mean of 4517 stock units in 1996-1997 (Figure 7).

Figure 7: Stocking Rate - Means of Quartiles by Year

Average stocking rates have decreased between 5% and 8% from 1976 to 1997 (Figure 6).

From the land development incentives of the late 1970s through to 1995, farms performing in the bottom quartile generally had with the lowest stocking rate (Figure 7).

Raising more stock units has been possible because farmers have expanded their area, by an average of 15 % over the whole period.

As farms have expanded their total stock units, more work has been done by fewer paid employees. In 1976-77, each labour unit managed 2240 stock units. By 1996-97, each labour unit was responsible for 2730 stock units, an increase of 22% (Figure 8).

Figure 8: Stock Units per Labour Unit

Associated with the change in total stock numbers and less labour managing more land and more stock, there has been a decline since the late 1980s in the proportion of sheep to total livestock (Figure 9). In 1983 sheep represented 72% of total stock units, but by 1997 they comprise only 56%.

Figure 9: Sheep Stock Units - Proportion of Total - Mean of Quartiles by Year

From 1978 to 1996, farms performing in the top quartile have consistently had a lower proportion of the total stock units as sheep, and more cattle than farmers in other quartiles (Figures 9 and 10).

Figure 10: Cattle Stock Units - Proportion of Total - Means of Quartiles by Year

While the wish to reduce labour inputs may have influenced changes in the proportion of sheep and cattle farmed, the relatively low profitability of beef farming through the 1970s into the mid 1980s reinforced this trend. During the late 1990s, anticipation of a cyclical upturn in beef returns, a shift from cow-based to more profitable dairy-beef systems, and wider adoption of intensive grazing practices for of intensive beef finishing systems all helped maintain the shift in favour of cattle.

The area planted in forestry has increased by an average of 250% since 1983-84. However, the area planted in 1996-97 still only averaged 13 ha or 3.6% of the effective farm area. On farms in the lower quartile where stocking rates are lower, forestry now accounts for more than 4% of the effective farmed area (Figure 11).

Figure 11: Area in Forestry as a percentage of Effective Farm Area

Income Generation and the Cost of Production

Over the 21 year period gross livestock revenue increased almost three fold to an average of $37.20/su in 1996-97. Throughout most of that time, gross livestock earnings have been closely associated with quartile rankings (Figure 12).

Figure 12: Gross Livestock Revenue per s.u.

The peaks in livestock earnings were in 1984-85 when a good production season was combined with high sheep meat, wool and beef returns (NZMWBES, 1992), and in 1993-94 which resulted from another good season and a peak in beef returns (NZMWBES, 1999).

By contrast, the low in 1994-95 and 1995-96 reflected declining beef returns, compounded by the previous period of expansion in the proportion of cattle farmed, and a weak market for lamb.

Despite the variable emphasis that farmers have placed on wool, high wool production has consistently been associated with the level of disposable profit of farm businesses (Figure 13).

Figure 13: Weight of Wool Sold per Sheep s.u.

Since 1993-94, there appears to be a trend to smaller differences between quartile mean wool production. The reasons are unclear but may in part reflect a response to the trend in fertiliser expenditure through the 1980s and 1990s (Figures 14 and 15).

Figure 14: Nominal Fertiliser Expenditure - $/s.u. by Quartile

Figure 15: Fertiliser Expenditure per su. - % of Mean - Lower and Upper Halves by Year

By June 1989, during the 1988-89 East Coast drought, top quartile farmers on average destocked by 14% compared to farmers the other quartiles where the average decrease was only 2% (Figure 16).

Figure 16: Shifts in the Proportion of Total Livestock Farmed

The gross revenue per labour unit increased 350% from 1976-77 to 1996-97 as a result of improved efficiency in labour use and changes in the revenue per stock unit (Figure 17).

Figure 17: Gross Revenue per Labour Unit

Since 1977-78, the cost of maintaining animal health has doubled to a level equivalent to 5-6% of 1997 gross farm revenue (Figure 18). Some new technology, such as faecal egg counting, may have enabled more efficient and targeted use of resources. However much of the increase in cost results from relatively more expensive new technologies such as pour-on internal and ecto-parasite control, new generation anthelmentics, and the wider use of new vaccine technologies.

Figure 18: Animal Health Expenditure as a Proportion of GFR.

While top quartile farms had relatively lower expenditure on animal health compared to gross livestock revenue, expenditure per stock unit in 1995-96 and 1996-97 at $2.00 per stock unit was 5% higher than the average for the other quartiles.

Expenditure associated with maintaining the assets of the farm business accounted for 18% of gross farm revenue throughout the period 1992-97. This contrasts with the period from 1985 to 1992 when farmers significantly reduced expenditure, and the period from 1978 to 1984 when development was encouraged and investment was focused on increasing productivity (Figure 19).

Figure 19: Weed, Pest, Fertiliser, Lime, Seeds, and R&M as a Proportion of GFR.

Expenditure on weed control was minimal during the late 1980s (Figure 20) in comparison with both the variability and higher expenditure that reflected the level of land development in the late 1970s and early 1980s. Since 1990-91, expenditure increased from the equivalent of 0.6% of gross farm revenue to 1.0% in 1996-97, in part reflecting increased thistle control in pastures opened up as a result of either drought or wet winter and spring pasture damage, and in response to cyclical patterns of porina infestation and the relative affordability of new control technology.

Figure 20: Expenditure on Weed and Pest Control as a Proportion of GFR.

Since the mid 1980s, repairs and maintenance expenditure has consistently averaged around 6% of gross farm revenue for each quartile group (Figure 21). Bottom quartile farms have fairly consistently spent a higher proportion of income in this category.

Figure 21: Expenditure on Repairs and Maintenance as a Proportion of GFR

Expenditure on maintaining pastures has generally accounted for half the total expenditure on asset maintenance. From the low level of the late 1980s, expenditure in the mid 1990s equated to an average of 11% of gross farm revenue or $3 - $4 per stock unit (Figure 14).

Through the decade from 1983 top quartile farms consistently spent the most per stock unit on fertiliser. Following the large rise in expenditure in 1993-94, bottom quartile farms maintained the momentum and through to 1997 had the highest average level of expenditure.

Change in the relative level of expenditure on fertiliser is a feature of the way in which farmers have responded to the changing fortunes of the industry. From 1981 to 1992 farmers in the lower two quartiles consistently spent at a below-average on fertiliser (Figure 15). Since 1992, this trend has reversed, in part as these farmers recognise the impact and constraint that soil fertility levels have on productivity. The prolonged period of below average expenditure may indicate that an extended period of above average expenditure is now required for a significant proportion of North Island Hill Country farm businesses if productivity is to be restored.

The expenses incurred in generating output account for an average of 50-53% of gross farm revenue for the period 1993-97. Since 1981, farmers in the bottom quartile have spent proportionately the most, and top quartile farmers the least, in generating income. The level of expenditure on Farm Working Expenses particularly differentiates the bottom quartile (Figure 22).

Figure 22: Farm Working Expenses - as a Proportion of GFR by Quartile

Cost of Business

Throughout the whole period, the cost of interest has been a significant feature of business for farmers in the bottom quartile, and is directly associated with the level of disposable profit for each quartile group (Figures 23 and 24).

Figure 23: Nominal Cost of Interest - $ per s.u. by Quartile

Figure 24: Cost of Interest - as a Proportion of GFR by Quartile

While, on average, bottom quartile farms have had the highest expenditure on interest through the 1990s, this is also the period in which farmers in this group made the most significant expansion in the area planted in forestry (Figure 11).

Standing charges have increased from 1976 to the late 1980s, consuming an increasing proportion of gross farm revenue. Since 1988 around 9% of gross farm revenue has been required to meet these service and local government charges (Figure 25).

Figure 25: Expenditure on Standing Charges (excl. interest) - as a Proportion of GFR by Quartile

Household

Personal drawings in 1996-97 averaged 22% of gross farm revenue, little different from the average of 21.5% in 1976-77. While the needs of the household may be similar, the larger enterprises associated with top quartile farms is in part reflected in the lower proportion of gross farm revenue appropriated for household needs (Figures 26 and 27).

Figure 26: Level of Drawings - as a Proportion of GFR by Quartile

Figure 27: Nominal Level of Drawings - $/s.u

The proportion of revenue appropriated for drawings is consistent with expenditure on farm working expenses. The lower quartile farmers spend the most and top quartile farmers the least.

As farm families sought to meet their needs through periods of changing profitability, other sources of income have been secured. Since the mid 1980s there has been a significant difference between quartile groups in the proportion of income from alternative sources detailed in farm business accounts (Figure 28).

Figure 28: Non-Farm and Other Sources of Income - as a Proportion of GFR.

Households within the bottom quartile group have on average had double the amount of off farm income compared to other groups.

While farm businesses in the bottom quartile have on average relied more on off farm income to meet these needs, they have also been more inclined to borrow to fund these needs. This contrasts with the position of businesses in the top quartile which, since 1983 have on average consistently repaid principal every year (Figure 29).

Figure 29: Net Annual Mortgage Reduction - as a Proportion of GFR.

The net result is a significant variation in the net liquidity of farm businesses across each of the quartile groups (Figure 30). One of the implications, especially for those businesses in the bottom quartile, is the relative inflexibility they have of meeting optimal production needs within the constraints of seasonal finance.

Figure 30: Level of Liquidity - Mean of Quartiles by Year

The relationship between farm Capital Value and gross farm revenue (Figure 31) highlights the periods of relatively inflated land prices in the early 1980s and mid 1990s.

Figure 31: Relationship of Capital Value of Land to Gross Farm Revenue by Year

The consistent relationship between each of the quartiles suggests that either techniques for assessing farm value fail to adequately differentiate between-farm advantages, or that the personal ability of the farmer could be better recognised.

A characteristic of businesses in the lower quartile is their association with relatively more capital invested in vehicles and plant (Figure 32).

Figure 32: Capital Value of Vehicles and Plant - Closing Value as a Proportion of GFR

Since the low level of the late 1980s, there appears to have been a trend to rebuild investments and other assets. While this appears to have been a consistent strategy of top quartile farmers, its recent association with bottom quartile farmers is noteworthy (Figure 33).

Figure 33: Investments and Other Assets - as a Proportion of GFR

During the 1990s there has been a proportionate increase in the value of the farm homestead (Figure 34). In part, this probably reflects a level of modernisation and refurbishment deferred throughout the previous decade.

Figure 34: Capital Value of Homestead - as a Proportion of GFR

The ratio of liabilities to gross farm revenue changed little across the two decades. This confirms a sense of equilibrium between profitability and borrowing (Figure 35). There is a suggestion that bottom quartile businesses may now have a higher ratio of liabilities to gross farm income than in the late 1970s. Inevitably this trend is limited by those farmers with a non-viable ratio leaving the industry but new owners replacing them.

Figure 35: Total Liabilities - as a Proportion of Gross Farm Revenue

Factors Contributing to Variation in Economic Performance

The financial viability of agriculture is closely linked to the efficiency with which farmers run their farm business.

The NZ Meat and Wool Boards' Economic Services has published analyses which "have consistently found the more profitable farms had higher stock performance and higher fertiliser usage. Good equity levels are found to assist profitability and increasing farm size is found to have a correlation with profitability and performance." (NZMWBES, 1997)

For this report, the 250 farms in the MWESNZ database for the most recent year available (1996-97) were split into quartiles based on a range of economic performance factors.

The analysis demonstrates the very wide range in the economic performance of North Island hill farms (Table 3). It also demonstrates that because the various parameters introduce different criteria which influence the placement of individual farms, each of the quartiles exhibits different characteristics. For example, EFS is strongly influenced by farm size, EFS per ha is influenced by land productivity and Rate of Return is influenced by capital value in relation to farm productivity.

Table 3: Quartile Average Performances for Farms When Grouped Using Each of Five Economic Parameters

Economic Parameter		Quartile
		Top	2	3	Bottom	All Farms
Farm Profit Before Tax	Mean	$112,823	$44,242	$20,980	-$46,561	$42,679
	Index	264	104	49		100
	Median	$89,942	$43,624	$20,505	-$2,353	$31,100
	Index	289	140	66		100
Economic Farm Surplus	Mean	$97,249	$29,571	$3,700	-$26,195	$25,886
	Index	376	114	14	-101	100
	Median	$74,956	$28,827	$4,089	-$23,134	$16,878
	Index	444	171	24	-137	100
Rate of Return	Mean	5.0%	2.4%	0.3%	-2.5%	1.4%
	Index	357	171	21	-179	100
	Median	4.6%	2.4%	0.3%	-2.3%	1.4%
	Index	329	171	21	-164	100
EFS/Effective ha.	Mean	$143.20	$67.45	$9.28	-$104.88	$54.28
	Index	264	124	17	-193	100
	Median	$137.89	$67.23	$11.35	-$97.77	$44.59
	Index	309	151	25	-219	100
EFS/Stock Unit	Mean	$14.38	$7.40	$0.98	-$10.68	$5.82
	Index	247	127	17	-184	100
	Median	$13.81	$7.24	$1.03	-$9.87	$4.55
	Index	304	159	23	-217	100

Notes:

(1) The "Index" shows the percentage relationship of each quartile to the "All Farms" average for either mean or median performance.

(2) Economic Farm Surplus (EFS) is an effective measure of the revenue generation and cost control with which farms are operate (Shadbolt and Bryant, 1999). Internationally, it is more commonly called "Operating Profit" and in Australia "Profit at Full Equity". EFS excludes debt servicing, tax, drawings and capital expenditure and therefore focuses on the generation of true profit. It is widely used in New Zealand to compare the economic efficiency with which farms are operated by indexing it per hectare, per stock unit and per $100 farm capital (in which case it is known as Return on Capital or Rate of Return (ROR)).

(3) Farm Profit Before Tax (FPBT) and EFS use the same calculation of revenue, farm working expenses and depreciation but differ in the other items subtracted: interest, rent and farm managers' salaries paid are subtracted in calculating FPBT whereas, but not in calculating EFS, although Wage of Management (WOM) is subtracted. The narrower range of "Index" values for FPBT results from the fact that the mean of all farms is higher than for EFS, because the average WOM is greater than the sum of average interest, rent and managerial salaries.

(4) For Farm Profit before Tax (FPBT) and Economic Farm Surplus (EFS), the mean of the quartiles equals the mean of all farms. However, the other three parameters are indices and, for them, group means are calculated as _  dividends ÷ _divisors. The widely different values occurring in the "Index" rows for the economic factors indicate that each of the economic factors has a quite different distribution pattern. The difference between the median and the mean indicates of the shape of the distribution for each quartile; where the two are close the distribution is relatively even across the range.

To further examine those factors contributing to the differences in economic performance between farms, the data were also analysed using two previously noted parameters:

• economic farm surplus, and
• rate of return

Physical Characteristics of Farms - Quartiles grouped by EFS

Grouping farms into quartiles according to EFS highlights the significance of farm size on profitability. From the top through to bottom quartile, farms get smaller in area but the difference between them gets progressively smaller too (Table 4). The top quartile is so much bigger than the rest that the mean of all farms is higher than the second quartile.

Table 4: Physical Characteristics of Farms - Means for Quartiles Grouped by Economic Farm Surplus

Farm size is an important factor in determining EFS. As some costs (e.g. electricity and administration) are relatively independent of the size of the farm, differences in the components of the EFS are largely a reflection of the difference in the size of the enterprise.

The pattern for stock numbers (expressed both as total stock units and capital value of stock) is similar to that for farm size. However there is no significant trend in stocking rates, the bottom quartile has the lowest rate, third quartile the highest. Top quartile farms run the highest proportion of sheep, the bottom quartile the lowest. Given the greater profitability of sheep over cattle in the year examined, an association between the percentage of sheep stock units and EFS is to be expected, but it is still weak. This contrasts with MWESNZ (1997) which showed, for the years 1991-95, that "high profit farms had much greater reliance on cattle than did the lower profit farms", and reflects the difference in relative profitability of sheep and cattle between the two periods.

The bottom quartile farms have the smallest area in forestry and farmers in quartiles 2 and 3 have a greater proportion of their farms in forestry than do top and bottom quartile farmers.

In terms of age of farmer, there is a moderately strong association with EFS quartile number. The average age of top quartile farmers is 47 and for bottom quartile farmers, 52.

Despite the common perception that small farms have an important role as stepping stones for new young farmers, the data show that small farms tend to be held by older farmers. There may be several explanations for this. They may have been held by the same owner for twenty years or more and failed to expand. In other cases they may be succession or semi-retirement blocks. Whatever the explanation, they are less used by young people and as small farms are operated at relatively low levels of efficiency opens to question their value as stepping stones into agriculture. Indeed their greater use may be as semi-retirement blocks, releasing the larger units to younger farmers.

Predictably, there are more labour units per property on the larger top quartile farms, but a very strong negative association between EFS quartile and stock units per labour unit (b = -533 s.u./labour unit). Part of this may be explained through the association of EFS with larger farms where labour (including the owner-operator's) is used more efficiently. This is in spite of the more profitable farms running a higher ratio of sheep which require more labour than cattle. More profitable farms also use more fertiliser per ha than do less profitable ones, but solely because they have higher stocking rates. There is little variation between quartiles in fertiliser use per stock unit.

More profitable, larger farms with higher sheep ratios naturally produce more wool. However, these higher EFS farms also produce more wool per sheep stock unit (b = 0.46 kg/s.s.u., r² = 0.967). This supports the observation by NZMWBES (1996) that the amount of wool sold per sheep increases with farm size (i.e. size in terms of GFR). With the extent of eight monthly shearing now done, the quartile into which some farms are placed may be determined in part by whether they sell wool from one shearing or two within the year examined.

Physical Characteristics of Farms - Quartiles grouped by Rate of Return

The impact of farm size on allocation to quartiles is reduced by using the data produced when farms are allocated to quartiles according to a size-related index. The options considered were EFS/ha, EFS/stock unit EFS/labour unit and Rate of Return (ROR).

ROR was chosen because it puts farms in the full range of sizes, physical environments and land types on a similar basis, and is an attempt to remove direct size effects. The only significant distortion created is that it undervalues the return for farms with a high capital value for reasons other than their productivity as sheep and beef farms (principally their suitability for conversion to other land uses or their desirable location). If EFS/ha were used, more extensive farms would be allocated to lower quartiles. Using EFS/stock unit, less well developed farms would be allocated to lower quartiles and if EFS/labour unit were used, more intensively managed farms would be allocated to lower quartiles.

Despite using Rate of Return to group the quartiles, there remains a strong size relationship across the quartiles (Table 5).

Table 5: Physical Characteristics of Farms - Means for Quartiles Grouped by Rate of Return

When divided by a factor such as capital value, some expenditure items which are independent of farm size still favour larger farms. These items include electricity, most administrative expenses, vehicle overheads, and a portion of management wages.

Total stock units have a relationship with farm size. Compared with the EFS-based analysis, there is a greater tendency for more efficient farms to run a higher proportion of sheep than less efficient farms. This is a further reflection of the relative profitability of sheep and cattle in the mid-1990s. This relationship to profitability has persisted since 1995 and the gross margin for sheep was higher than for cattle from 1995-96 to 1997-98 (NZMWBES Farm Surveys). However, as discussed elsewhere in this report, it was not until 1996-97 that top quartile farms ran a higher proportion of sheep than lower quartile farms. The reason for the association between high Disposable Profit and high cattle ratio in spite of low cattle gross margins 1995-96 is not known, although it is notable that in this first full year of low cattle profits Classes 3 and 4 farmers actually increased their cattle to sheep ratios.

Bottom ROR quartile farms have an even lower stocking rate and top quartile farms a higher stocking rate than in the EFS quartile analysis. Stocking rates fall almost twice as steeply across the ROR-based quartiles as across the EFS-based quartiles, showing that profitability is highly dependant on stocking rate, even though there is still a poor relationship between stocking rate and quartile number. On the other hand, there is a strong relationship between stock units per labour unit and ROR quartile number (as there is with EFS quartile number). Productivity of labour in terms of the number of stock per person has a very strong, probably causal relationship, with economic efficiency.

Total wool sold and wool sold per sheep stock unit are even more strongly associated with ROR quartile number than for EFS quartile number, (even though the differences are no larger) further reinforcing the strong association between livestock performance and economic efficiency.

Physical Characteristics of Farms per $Million Farming Capital - Quartiles grouped by Rate of Return

When quartiles are defined with ROR and expressed on a per-farm basis, there is still a very strong relationship between farm size and the distribution of farms by quartile. To further reduce this effect, the same farm data (i.e. allocated to quartiles by ROR) were expressed per million dollars of farm capital. This produced an EFS for each quartile which is the percent ROR multiplied by $10,000. However, this technique does not completely remove the effect of size. When expressed on this basis more profitable farms are still 40.5 ha larger than the next most profitable quartile and run 441 more stock units (Table 6).

Table 6: Physical Characteristics of Farms per $m Farming Capital - Means for Quartiles Grouped by Rate of Return

When expressed on a per $million farming capital basis, more profitable farms are still larger, run many more stock and sell much more wool than less profitable farms. This is achieved with less labour but more fertiliser. Top quartile farms run 1,282 (49%) more stock units per $million farming capital than do bottom quartile farmers. This is entirely accounted for by the 49% higher capital value per stock unit of bottom quartile farms ($382/s.u) compared to top quartile businesses ($256/s.u).

Forestry area, percentage of forest per farm and per $million farming capital are greater for quartiles 2 and 3 than for the top and bottom quartiles. Possibly the most profitable farmers are less interested in longer term investment, and the least profitable farmers believe they cannot afford the capital, the time, or the loss of grazing to set aside land for forestry.

Comparison of the Analyses

Table 7 compares the regression coefficients of the physical characteristics with quartile number for each of the three analyses. Because the same farms are in the quartiles for both the ROR-based analyses, the coefficients for farmers' age are the same, and all indices are similar whether expressed per farm, or per $million farming capital.

Table 7: Regression Coefficients of Quartile Farm Features

Equity in the Farm Business and Home

Issues of farm equity were examined. These include the assets and debt associated with both the farm business and the residence of the farmer. As used here, equity is taken to include off-farm assets and investments, so could be called "net worth". Opening assets were used except for homesteads and cars, for which only closing values were available.

Equity in the Farm Business and Home - Quartiles grouped by EFS

There is a strong positive association between quartile number and farm capital per stock unit. Almost all of this association is attributable to how the value of land and buildings per stock unit vary across the quartiles. A high EFS is therefore strongly associated with the low capital value of land and buildings per stock unit, as well as large farms (Table 8).

Table 8: Farm/home Equity and Its Components - Means for Quartiles Grouped by Economic Farm Surplus

The low value of land and buildings per stock unit has very little link with stocking rate, and the association of EFS with capital value/ha is almost as strong as with Capital Value/s.u. This negative association of profitability with land value may result from the association with farm size. Large farms may have lower values per hectare and per stock unit because they have fewer buyers competing for them and are often more remote from opportunities for off-farm employment, population centres, schools and other services.

The capital value of land, farm buildings and homesteads has a similar relationship to land area as does total farm capital. Although the regression coefficient of total farm capital on quartile number is $294,000, the difference between the top and second quartile is $836,800 highlighting just how much bigger the top quartile is than any other.

Non-farm capital comprises Income Equalisation Scheme deposits, current assets, term deposits, investments, the homestead, car, and other assets. All of these items are associated negatively with EFS quartile number. The relationship of total non-farm capital to quartile number is not strong and of all these assets, car value has the strongest relationship. The main reason for the weakness of the relationship to these items is that bottom quartile farmers have higher levels of off-farm investment and other assets than have third quartile farmers. Low performing farmers (on small farms) appear to recognise the need to diversify their sources of income.

Level of fixed liabilities is strongly related to EFS quartile number, but current liabilities are less strongly associated. However, liabilities per hectare and per stock unit are weakly related to quartile number, reflecting a low level of association between debt per unit size and business efficiency.

Farm/home equity has only a moderate association with EFS quartile (r² = 0.566); the average top quartile farm has a high equity, the other three quartiles have lower but similar equity. The level of equity differentiates farms at the top end, but not the bottom.

Percentage equity, is positively associated with EFS quartile number, and is highest for bottom quartile farms. This surprising finding could be explained at least in part by two factors. First, the greater age of bottom quartile farmers means they have had longer to build up their equity, or that in purchasing a semi-retirement property, a significant portion of any borrowing is undertaken and serviced by the succeeding generation. Second, it is likely that more heavily indebted top quartile farmers strive harder to achieve and operate their farms as efficiently as possible so as to increase their level of financial security (updating the observation once made by the then Agriculture Minister, The Hon Duncan McIntyre in a reference to a farmer's effort being determined by the size of the monkey on his back). Overall, the positive association between equity and EFS demonstrates the current low level of relationship between farm efficiency and indebtedness.

Equity in the Farm Business and Home - Quartiles grouped by Rate of Return

The capital value of land per stock unit, vehicles and plant per stock unit and total farm capital per stock unit all have strong positive relationships with quartile number. This was noted on for EFS quartiles (Table 9).

Table 9: Farm/home Equity and Its Components - Means for Quartiles Grouped by Rate of Return

The large items - land, and livestock and fixed liabilities - contribute most of the variation in farm equity between ROR quartiles. However, for their size, off-farm investments and other assets have the most potent positive association, and reserve the most potent negative association with variation between quartiles. Arguably efficient farmers have been able to make the greatest use of off-farm investments and the greatest use of family assistance through livestock bailment and other assets than their less efficient counterparts.

There is almost no association between debt level per hectare or per stock unit and Rate of Return quartile number, confirming the already-demonstrated current lack of relationship between debt and profitability. The value of homesteads, on the other hand, is a weak contributor to variation. As homesteads are not included in the farm capital for the ROR calculation, it cannot be said that over-capitalisation with higher value homesteads contributes to lower profitability. The weak association simply indicates that the standard of the dwelling bears little relationship to the efficiency with which the farm is run. The same conclusion can also be drawn about the value of the car of the farm family.

Equity in the Farm Business and Home per $M Farming Capital - Quartiles Grouped by Rate of Return

Moving from the top to the bottom quartile, land and buildings and homesteads rise in value per $m farming capital and the other equity components fall (Table 10). "Other assets" is the most potent contributor to variation. Farms in the top two quartiles have much higher amounts of these. Interestingly, MWESNZ (1999) shows that Class 3 (NI Hard hill country) farmers have 85% more "Other Assets" than Class 4 (NI Hill country) farmers. The association of debt with quartile number is weak because Quartile 2 farmers have the highest debt and Quartile 3 farmers have the lowest debt per $million farming capital. A summary of the contributions made to variation between quartiles in farm/home equity by its components is shown in Tables 11 and 12).

Table 10: Farm/home Equity and Its Components per $m Farming Capital - Means for Quartiles Grouped by Rate of Return

	Quartile					Regression		R2
	Top	2	All	3	Bottom	Coefficient
Opening Capital
Land & buildings	759,299	795,793	798,280	824,187	832,573	24,821	0.9327
Vehicles & plant	20,539	20,443	20,411	19,513	21,387	161	0.0737
Livestock	220,162	183,764	181,310	156,301	146,041	(24,983)	0.9420
Farming Capital	1,000,000	1,000,000	1,000,000	1,000,000	1,000,000	0	1.0000
Current assets	45,397	29,635	35,611	28,093	37,802	(2,433)	0.1538
Off-fm Investment	65,815	38,762	41,527	14,369	41,519	(9,728)	0.3561
Other Assets	6,535	12,853	6,836	4,156	2,474	(2,088)	0.3507
Homestead (cl)	71,327	80,918	80,171	79,093	95,688	7,126	0.8120
Car (cl)	8,137	7,648	7,652	7,689	6,774	(405)	0.8349
Current Liabilities	37,562	39,743	34,210	28,695	28,221	(3,907)	0.7149
Fixed Liabilities	145,908	181,092	140,879	111,444	115,644	(16,044)	0.4115
Total Liabilities	183,470	220,835	175,089	140,139	143,865	(19,951)	0.4596
Reserves	41,541	57,381	37,999	28,302	17,729	(10,052)	0.5720
Farm/home Equity	972,201	891,600	958,708	964,961	1,022,664	22,475	0.2883

Table 11: Contribution of Items to Differences in Farm and Home Equity

Table 12: Interaction of Factors and Economic Parameters on Differences in Equity Between Quartiles

Farm Business Operation and Efficiency

Irrespective of the physical and financial structure of the farm, the effectiveness and efficiency of the operation of the farm business can be expected to directly impact on the performance of the farm business. It is this aspect of business management that is front and foremost in farmers' minds and the area which they are most able to influence quickly through strategic and day-to-day decision making.

Farm size has a dominant effect in differentiating the performance of top quartile businesses from the other three when quartiles are separated on EFS (Table 13). This differs from the findings of MWESNZ (1997) that "size is a determinant of performance only for very small farms".

Table 13: Farm Business Operation and Efficiency - Quartiles Grouped by EFS

When quartiles are separated on ROR, EFS is even more closely associated with quartile number than when separated on EFS itself; for every step from the top to the bottom quartile, it reduces by $38,055 (Table 14).

Table 14: Farm Business Operation and Efficiency - Quartiles Grouped by Rate of Return

	Quartile					Regression		r2
	Top	2	All	3	Bottom		Coefficient
Revenue
Cash Livestock Revenue	270,093	180,718	168,214	142,934	80,329		(60,708)		0.9751
Value of Livestock Change	6,858	1,543	(1,809)	(7,916)	(7,682)		(5,308)		0.8908
Livestock Revenue	276,951	182,261	166,405	135,018	72,647		(66,016)		0.9795
Grazing	5,991	4,624	4,322	3,520	3,167		(958)		0.9446
Other Revenue	5,684	4,754	4,706	4,074	4,319		(478)		0.7560
Gross Farm Revenue	288,627	191,639	175,434	142,612	80,132		(67,451)		0.9791
Expenses
Animal Health	14,495	9,719	8,887	7,372	4,026		(3,375)		0.9811
Weed & Pest	1,744	2,057	1,772	1,487	1,798		(41)		0.0508
Fertiliser	26,539	20,820	18,753	16,182	11,556		(4,959)		0.9971
Seeds & Lime	1,685	1,026	1,293	973	1,486		(65)		0.0579
Repairs & Maintenance	12,961	9,824	9,075	7,863	5,694		(2,376)		0.9884
Other Farm Working Expenses	70,012	50,472	48,053	43,561	28,443		(13,161)		0.9702
Farm Working Expenses	127,436	93,918	87,833	77,438	53,003		(23,978)		0.9823
Rates & Insurance	12,195	9,973	9,116	8,100	6,227		(1,978)		0.9981
Cash EFS Expenses	139,631	103,891	96,949	85,538	59,230		(25,955)		0.9841
Depreciation	11,483	10,395	9,449	9,235	6,712		(1,547)		0.9524
Wages of Management	46,556	42,633	43,150	43,524	39,947		(1,894)		0.8052
Total EFS Expenses	197,670	156,919	149,548	138,297	105,889		(29,396)		0.9814
EFS	90,957	34,720	25,886	4,315	(25,757)		(38,055)		0.9726

When farm size effects are minimised by working on a per $million farming capital basis, the same seven items as in the per-farm analyses are still the most important, but the order is changed. The wage of management increases as EFS falls across the quartiles. This is to be expected as wage of management is partly per-farm based and there are fewer farms per $million farming capital in the top quartiles. Weed and pest control and lime and seeds also change but there is only a weak association between higher expenditure on these items and less profitable farming (Table 15).

Table 15: Farm Business Operation and Efficiency per $M Farming Capital- Quartiles Grouped by Rate of Return

Expenditure on animal health is the most potent item of expenditure. Every $1 increase in animal health expenditure across the quartiles is associated with a $17.57 increase in EFS. Fertiliser expenditure makes about the same contribution to EFS variation as animal health, even though on the average farm it is more than double the proportion of expenditure on animal health expenditure.

Comparison of the Analyses

Revenue from livestock and livestock products is by far the largest determinant of EFS. Gross farm revenue is twice as important in determining EFS as all expenditure items combined (Table 16). This confirms the emphasis placed on farm production by MWESNZ in many of their publications. "Other Farm Working Expenses" is second, and fertiliser is the third most important group. Value of livestock increase is next in importance. The top two quartiles increased livestock values in this year (1996-97) whereas the bottom two quartiles reduced livestock values. Animal health and repairs and maintenance follow in importance.

Table 16: Interaction of Factors and Economic Parameters on Differences in EFS between Quartiles

In relation to their size as contributors to EFS, livestock cash revenue, animal health expenditure and, possibly, value of livestock change are the most influential. The strong association between animal health expenditure and EFS is significant. While it may be unwise to suggest the relationship is a direct cause-and-effect one, the point can be made that successful farmers do not stint on animal health expenditure and they are well rewarded for this behaviour. On the other hand, wage of management has little influence for its size, and lime and seeds, weed and pest control, and other (non-livestock) revenue are both small in size and lack influence on EFS.

Sources and Application of Funds - Quartiles Grouped by EFS

The cash component of EFS provides the bulk of variation in the sources of income recorded in farm accounts by quartile. "Other Sources", and interest and dividends significantly reduce the variation in farm income (Table 17).

Table 17: Sources and Application of Funds - Quartiles Grouped by EFS

Sources and Application of Funds per $Million Farming Capital - Quartiles Grouped by Rate of Return

In all analyses of the sources of funds, the EFS cash component dominates a reducing trend across the quartiles. While all the other items counteract this, their impact is inconsistent. In all analyses, Quartile 2 receives more interest and dividends and other non-farm income than does Quartile 3. The distribution of non-farm income between quartiles varies widely and inconsistently and the reasons for this remains unclear. In the analysis per $million farming capital, the regression coefficient for total sources (and applications) of funds is reduced to only $8,162 (Table 18). In effect, less efficient farmers appear remarkably effective in generating the cash to meet the shortfall produced by their farms' low earnings.

Table 18: Sources and Application of Funds per $Million Farming Capital - Quartiles Grouped by Rate of Return

Once the effects of farm size are minimised, regression coefficients change from negative to positive for some items - namely mortgage reduction, new buildings and new vehicles and plant. Thus, less profitable farms, while spending generally less per farm on these items, spend slightly more per $million farming capital. As circumstances differ from farm to farm, it can only be speculated whether the higher rate of new buildings, vehicles and plant are justifiable deferred expenditure for many low-income farms or best described as over-capitalisation.

Sources of Funds - Comparison of Analyses

It might be expected that reducing or removing farm size effects would reduce the trend for lower quartile farms to rely on off-farm sources of income. However, in the analysis of ROR per $m farming capital, the influence of the EFS cash component is reduced and that of other off-farm sources increased (Tables 18 and 19). It would seem therefore, that less profitable farms do not place greater reliance on non-farm sources of income because they are smaller, but for some other reason or reasons. From the top to the bottom quartile in both per farm based analyses, each quartile group has less money available. In the EFS-based analysis, bottom quartile farms have only slightly less funds available than third quartile farms (Table 17). However, in the ROR analysis moving from the top to the bottom ROR quartile, there is a relatively even reduction in the amount of money available as one moves from the top to bottom quartiles (Table 20).

Table 19: Contributions of Items to Variation in Sources of Funds

Table 20: Sources and Application of Funds - Quartiles Grouped by Rate of Return

Application of Funds

For both EFS-based and ROR-based analyses, interest and drawings are most closely associated, although negatively, with quartile number i.e. the less efficient the farm, the higher the interest and drawings (Tables 17, 18, 20, 21 and 22). The effect of farm size again dominates. Drawings, interest and tax, between them make up less than 57% of the total application of funds yet are responsible for 68% to 70% of the variation observed between quartile numbers. Tax, interest and "Other applications" are the most potent sources of variation. While tax makes up only about 7% of the average funds application, it contributes at least 15% of the differences between quartiles in funds allocated.

Table 21: Contributions of Factors to Variation in Application of Funds

Table 22: Interaction of Factors and Economic Parameters on Differences in Application of Funds between Quartiles

Change in Equity - Quartiles Grouped by EFS

Equity per farm decreased during the year by an average of $35,936 (Tables 23 and 24). The value of land and buildings fell across all quartiles by an average of $46,601, while the decline in livestock value was averaged only $1,809. These reductions were only partly offset by increases in working capital, vehicles and plant and "Other Capital", and reductions in fixed liabilities and reserves.

Table 23: Components Contributing to Change in Equity - Quartiles Grouped by EFS

Table 24: Components Contributing to Change in Equity - Quartiles Grouped by Rate of Return

Change in Equity - Comparison of Analyses

Lower quartile businesses have invested relatively more in changing vehicles and plant during the year that upper quartile groups ($2,076 and $256 cf. -$426 and -$6). By contrast, upper quartile groups continued to invest in livestock while their peers reduced their investment ($4,011 and $1,098 cf. -$5,722 and -$7,792) (Table 25). In general, therefore, top quartile farmers appear to have made decisions on capital changes which increased productivity or consolidate their financial positions whereas lower quartile farmers tended to have acted less positively. However, lack of information on homestead and car changes, limit any firm conclusions.

Table 25: Components Contributing to Change in Equity per $Million Farming Capital - Quartiles Grouped by Rate of Return

(Tables 26 and 27) rank the contribution made to the variation between quartiles in change in equity over the year by each item.

Table 26: Contributions of Factors to Variation in Change in Equity

Table 27: Interaction of Factors and Economic Parameters on Change in Equity between Quartiles

Table 28: Financial Indicators: Revenue and Net Operating Profit After Tax - Quartiles Grouped by Rate of Return

Table 29: Value Created - Quartiles Grouped by Rate of Return

Notes:

Every quartile in both analyses produced a negative Value Created. The value created by the top quartiles, at -$5,370 and -$10,411 (for EFS and ROR based groupings respectively) suggests that only a very small proportion of surveyed farmers achieved a positive value created. The majority are effectively mining the capital of their business. According to this business measure, farming is in a poor economic state.

Taking the bold step of eliminating farm debt, so that all capital is costed as equity capital, would put the top EFS-based quartile into a positive value creating situation but the other groups would still be in a negative situation, as would all the quartiles in the ROR-based analysis.

Contact for Enquiries

Manager
North Island Regions
Sector Performance Policy
MAF Policy
Hamilton
NEW ZEALAND

Phone: +64 7 957 8313
Contact this person