Regional and national impacts of the 2007-2009 drought
Prepared for MAF by Butcher Partners Ltd
July 2009
Download full report as a PDF [462KB PDF]
Appendix 3: Detailed worksheets showing changes in expenditure by farm types, and sources of economic impacts by region
Download Appendix 3 Tables as excel files:
1. Executive summary
A widespread drought affected most of New Zealand from late 2007 until autumn 2008 with the impact covering two farming seasons. The only regions not to be significantly affected were Northland, Gisborne and Westland. MAF Policy commissioned Butcher Partners Ltd, with the assistance of The Agribusiness Group, to estimate the economic impact of the drought, and the regional distribution of this impact.
The primary method of direct impact analysis was to compare forecast budgets with actual outcomes for relevant farm types in each affected region. The forecasts were made prior to the drought and reflect a “without” drought expectation of physical performance and farming policy. The actual outcomes reflect a “with drought” situation. The difference between the two reflects the direct on-farm impacts of the drought1.
As is shown in Table 1.1, output declined by $1.49 billion over the two-year period of the drought. The impacts of the drought on output are a combination of the difference in farm gate sales and the difference in changes in capital livestock2. Farm value added (net income) declined by $1.89 billion over the two-year period of the drought. This arose from a combination of the decline in output and an increase in the cost of inputs. There was a further loss of value added associated with a run-down in reserves of on-farm feed, and this was probably about $0.1–0.2 billion. There was little discernible change in direct farm employment or household income.
Table 1.1: Cost of the drought by farm type
| 2007/08 and 2008/09 combined | Dairy | Sheep and Beef | Deer | Total |
|---|---|---|---|---|
| Loss of income ($M) | (927) | (262) | (1) | (1 190) |
| Change in Capital Stock ($M) | - | (303) | - | (303) |
| Decline in output | (927) | (565) | (1) | (1 493) |
| Expenditure Impact ($M)* | (550) | 161 | (4) | (392) |
| Total Drought Cost ($M) (Loss in Value Added) |
(1 476) | (404) | (5) | (1 885) |
| Sector share of total cost | 78% | 21% | 1% | 100% |
| Cost as % of production in the two drought years | 8% | 4% |
*A negative value implies an increase in expenditure. Excludes run-down in on-farm stocks of feed.
To get some feel for the significance of the drought, we compared the loss of value added over the two drought years to expected non-drought farm income over those two years. We estimate that for dairy farms the losses due to the drought were equivalent to about 8 percent of production over the two years, while in the case of sheep and beef farms it was equivalent to about 4 percent of production.
These losses may not seem overly significant, but when expressed as a proportion of net farm income, which is the income available to cover depreciation, a return to the labour of the farm family via drawings and a return to capital, the decline is two to three times as great (in the order of 15 percent).
In financial terms, the impact of the drought was felt most heavily by dairy farmers, who suffered almost 80 percent of the loss in value added (see Table 1.2). This impact was focused in Waikato, which suffered 44 percent of the national costs of the drought.
Table 1.2: Cost of drought (direct value added) by location
| Dairy | Sheep & Beef | Deer | Total | Share | |
|---|---|---|---|---|---|
| Northland | - | - | (0.0) | (0) | 0.0% |
| Auckland | (14) | 5 | (0.1) | (9) | 0.5% |
| Waikato | (799) | (37) | (0.6) | (836) | 44.4% |
| Bay of Plenty | (155) | (9) | (0.3) | (164) | 8.7% |
| Gisborne | (1) | - | (0.1 | (1) | 0.0% |
| Hawkes Bay | (14) | (61) | (0.5) | (76) | 4.0% |
| Taranaki | (181) | (26) | (0.1) | (207) | 11.0% |
| Wanganui/Manawatu | (91) | (133) | (0.5) | (225) | 11.9% |
| Wellington | (21) | (43) | (0.1) | (64) | 3.4% |
| Tasman | (17) | (2) | (0.1) | (19) | 1.0% |
| Marlborough | (1) | 1 | (0.1) | 0 | 0.0% |
| Canterbury | (35) | (6) | (1.6) | (43) | 2.3% |
| West Coast | (51) | (1) | (0.1) | (53) | 2.8% |
| Otago | (26) | (35) | (0.7) | (62) | 3.3% |
| Southland | (71) | (56) | (0.7) | (127) | 6.8% |
| New Zealand | (1 476) | (404) | (5.4) | (1 885) | 100.0% |
Flow-on impacts
Up until the mid-1990s, typical drought economic impact analysis in New Zealand related primarily to sheep and beef farms. Total impacts were calculated using average farming multipliers, which implicitly assumed that farm inputs declined by the same proportion as farm outputs. Hence there were calculated to be significant flow-on effects in the rest of the economy. On sheep and beef farms in severe and prolonged drought or when farmers are in a weak financial position, this tends to be the case, as is borne out by casual observation and anecdotal information. However, the direction of expenditure shift is less certain where the drought affects dairy farms that are less able to change inputs without severely affecting output in the short-term, are not used to coping with drought, or when the drought occurs at a time when farmers are in a strong financial position.
In the case of the 2007/08 drought, dairy farmers were getting exceptionally high prices for milksolids. Their response was to spend heavily on feed to maintain production as much as possible and then on pasture resowing to restore lost production as quickly as possible after the event. Sheep and beef farmers had already been financially squeezed for some time with low profitability and had limited opportunity to reduce costs further. Sheep and beef farmers sold capital stock in 2007/08 hence improving cash flow. In 2008/09 they enjoyed a substantial improvement in product prices, but did not restock to the same extent as they destocked, hence again enabling them to maintain expenditure in other areas. For all of these reasons the drought led to an increase in farm spending.
To estimate the flow-on effects of changes in farm spending we estimated changes in farm spending by category of spending and by region. We allocated this spending to generate estimates of changes in production by industry by region3, and applied relevant industry regional and national multipliers to estimate total indirect and induced economic impacts associated with this change in spending.
Milk for processing declined significantly. We have converted this to changes in output of processed dairy products and applied dairy industry multipliers4 to estimate the regional economic impacts of this reduction. Meat processing increased in 2007/08 as farmers killed capital stock, but declined in 2008/09 as livestock production declined with a smaller base of breeding stock and poorer reproductive performance. We converted the change in the value of livestock going to slaughter to a change in the output of meat processing works and applied meat industry multipliers to estimate the total regional and national economic impacts of this.
We estimate that the total loss of off-farm value added as a result of the drought was $887 million (see Table 1.3). This large loss was predominantly due to the impacts of the drought on dairy processing (down in both years) and meat processing (up in 2007/08 and down in 2008/09). The negative impacts of this reduction in processing sector activity completely swamped the small positive effect associated with increased farm spending on such items as pasture restoration. It is also possible that the loss of off-farm value added is understated, because some costs are fixed overheads and the marginal value added: output ratios are higher than the average ratio.
Table 1.3: Direct, off-farm and total output and value added impacts of drought, by region ($M)
| Region | Output ($M) |
Value Added ($M) |
||||
|---|---|---|---|---|---|---|
| Direct | Off-farm | Total | Direct | Off-farm | Total | |
| Northland | -0 | -34 | -34 | -0 | -14 | -14 |
| Auckland | 1 | -132 | -130 | -9 | -76 | -85 |
| Waikato | -624 | -1 449 | -2 073 | -836 | -384 | -1 221 |
| Bay of Plenty | -107 | -230 | -337 | -164 | -68 | -232 |
| Gisborne | -0 | -3 | -4 | -1 | -2 | -3 |
| Hawkes Bay | -100 | -17 | -117 | -76 | -49 | -125 |
| Taranaki | -154 | -278 | -432 | -207 | -91 | -298 |
| Wanganui/Manawatu | -211 | -46 | -258 | -225 | -89 | -314 |
| Wellington | -62 | -40 | -102 | -64 | -41 | -105 |
| Tasman | -9 | -10 | -18 | -19 | -3 | -21 |
| Marlborough | -1 | -2 | -3 | 0 | -1 | -1 |
| Canterbury | -44 | 186 | 143 | -43 | 3 | -40 |
| West Coast | -19 | -21 | -40 | -53 | -6 | -59 |
| Otago | -55 | 15 | -40 | -62 | -27 | -89 |
| Southland | -109 | -6 | -115 | -127 | -38 | -165 |
| New Zealand | -1 493 | -2 068 | -3 561 | -1 885 | -887 | -2 773 |
We estimate that the drought led to a potential loss of almost 3000 job-years5 of work off-farm and a loss of $143 million of household income (see Table 1.4). This is equivalent to 1500 full time jobs being lost for the duration of the drought. Given that there are 9500 people employed in milk processing, 24 000 in meat processing and probably twice as many in the various support industries for these sectors, the numbers seem realistic.
In many cases the impacts on jobs will have been felt as shorter working periods in the seasonal industries or on fewer hours per day. It is also probable that our estimates of potential lost household income and employment overstate the actual impacts because average multipliers will not be the same as marginal multipliers. For example, tanker drivers may face only a small reduction in hours, even if they are picking up 10 percent less milk at each farm, and the processing sector will be reluctant to make staff redundant if they see that the downturn is likely to be short-lived. This means that business will not have been able to adjust completely to the reductions in throughput caused by the drought and will most probably have reduced productivity rather than reduced employment by the amount reported here.
The probable overstatement of lost household income off-farm is a partial offset to the likely understatement of lost household income on-farm. Anecdotal evidence is that even although the model farm budgets show little change in farm drawings, some families had a very severe drop in disposable income and household spending.
Table 1.4: Direct, off-farm and total employment and household income impacts of Drought, by region ($M)
| Region | Employment (Job-years) |
Gross Household Income ($M) |
||||
|---|---|---|---|---|---|---|
| Direct | Off-farm | Total | Direct | Off-farm | Total | |
| Northland | 0 | -100 | -100 | 0 | -4 | -4 |
| Auckland | 0 | -460 | -460 | 0 | -21 | -21 |
| Waikato | 0 | -2 728 | -2 728 | 0 | -126 | -126 |
| Bay of Plenty | 0 | -438 | -438 | 0 | -20 | -20 |
| Gisborne | 0 | -13 | -13 | 0 | -1 | -1 |
| Hawkes Bay | 0 | -67 | -67 | 0 | -3 | -3 |
| Taranaki | 0 | -395 | -395 | 0 | -18 | -18 |
| Wanganui/Manawatu | 0 | 106 | 106 | 0 | 1 | 1 |
| Wellington | 0 | -144 | -144 | 0 | -7 | -7 |
| Tasman | 0 | 20 | 20 | 0 | 0 | 0 |
| Marlborough | 0 | -4 | -4 | 0 | 0 | 0 |
| Canterbury | 0 | 818 | 818 | 0 | 36 | 36 |
| West Coast | 0 | 67 | 67 | 0 | 2 | 2 |
| Otago | 0 | 143 | 143 | 0 | 6 | 6 |
| Southland | 0 | 240 | 240 | 0 | 12 | 12 |
| New Zealand | 0 | -2 959 | -2 959 | 0 | -143 | -143 |
The estimates of economic impacts ignore any restocking in 2008/09, and any associated reduction in meat slaughtering. While comprehensive data is not yet available, initial indications are that many sheep and beef farmers have only partially restocked. Reasons include:
- Some land is still in drought and restocking is not appropriate.
- Some land has converted to other uses such as dairy grazing.
- Some farmers have decided to retain fewer capital stock but try and achieve higher reproduction rates and slaughter weights.
- Some farmers are simply waiting to see whether the recent upturn in stock prices will continue and whether it is worth remaining in sheep and beef farming.
Limitations of the study
While the process adopted for the analysis had its limitations with regard to accuracy, particularly in estimating results at a regional level, we were unable to identify any method of estimating drought impacts at a regional level other than to use the MAF Farm Monitoring models. While this study converted forecast model values to reflect changes in international commodity prices, no other parameter values were changed. Future analysis could consider re-running the “without drought” forecast budgets using the same non-drought-affected parameter values as were revealed by the “actual” budgets.
Advantages of the process used
- The inclusion of the benefits that accrue to those finishing farms that benefit from the purchase of store stock at very low prices, hence avoiding potential overstatement of drought costs.
- The only simple way of calculating regional impacts.
- Includes adjustments for sales of capital stock, hence avoiding understatement of drought impacts during the destocking phase and overstatement during a restocking phase.
- Avoids the need for one-off large-scale surveying of individual farm types by taking advantage of the existing on-going surveying that underlies the various farm models. The down-side is that specific questions directly relevant to the impacts of the drought are not asked. Impacts have to be inferred from data that might also incorporate other impacts.
1The forecast budgets were adjusted to reflect subsequent changes to parameter values that had not been affected by drought. For example, they were adjusted to reflect actual international prices rather than those that had been expected at the time of the forecasts.
2Generally sales fell as a result of the effects of the drought on production, but rose as a result of destocking. Sales of capital stock do not reflect changes in production, but reflect changes in ownership of the asset base.
3In some case to imports or to changes in stocks of feed, which have little flow-on impact apart from on the transport sector.
4In order to avoid double counting of on-farm economic impacts, we calculated modified dairy and meat processing multipliers, which excluded backward linkages through farms.
5A job-year is one person working full-time for one year; four persons working full-time for three months, etc.
Contact for Enquiries
Manager
North Island Regions
Sector Performance Policy
MAF Policy
Hamilton
NEW ZEALAND
Phone: +64 7 957 8313
Contact this person
