The private optimum

Table 9 presents the privately optimal steady state phosphorus fertilization and the STP levels, and the associated steady state outcomes: the crop yield, the profits and the phosphorus loss from each parcel type. The phosphorus loss is a sum of DRP-loss and the bioavailable fraction of the PP-loss. Each entry in the table is a value of the variable given in the leftmost column for the associated parcel type.

Table 8. The private optimum.

Slope 0.3 0.7 2 5 7

Shape 1:1 1:3 1:1 1:3 1:1 1:3 1:1 1:3 1:1 1:3

Fertilization (kg ha^-1) 20.4 20.4 20.4 20.4 20.4 20.4 20.4 20.4 20.4 20.4

STP 7.9 7.9 7.9 7.9 7.9 7.9 7.9 7.9 7.9 7.9

Crop yield (kg ha^-1) 3363 3363 3363 3363 3363 3363 3363 3363 3363 3363 Profits (€ ha^-1) 96.4 96.4 96.4 96.4 96.4 96.4 96.4 96.4 96.4 96.4 Phosphorus loss (kg ha^-1) 0.54 0.54 0.55 0.55 0.69 0.69 1.51 1.51 2.45 2.45

The annual steady state phosphorus fertilization is 20.4 kg ha^-1 for all parcels. This is associated with a steady state STP value of 7.9. The annual crop yield associated with these is 3363 kg ha^-1 and the annual profit equal to 73.5 € ha^-1. The information value of the steady state profit within this study is limited due to its stationary nature.

However, we report it in some cases, as well as the stationary crop yields.

The heterogeneity in our model (slope and shape) does not affect crop yields or costs in our model. Therefore, all parcels have identical privately optimal solutions, only the associated phosphorus losses are different. These differences, on the other hand, are substantial. The three most gentle parcels have phosphorus losses fairly close to each other. However, the phosphorus loss from the parcel with slope 2% is already about 28% higher than from the most gentle parcel. For the steepest parcel, the loss is more than four times higher than from the gentlest one. The differences stem from dissimilar erosion intensities between parcels with different slopes. The phosphorus loss from the whole target area is 36.4 kg.

In chapters 3 and 4, we conducted already the analytical comparative statics. Now, we will do a similar exercise numerically to analyze the magnitudes of the effects of parameter variation on optimal allocations (fertilization and STP). If the effects on outcomes (crop yields and profits) are particularly interesting we will discuss these as well.

Variation in prices

We calculate the optima using 5 different input prices: the default value, a 5%

decrease, a 10% decrease, a 5% increase and a 10% increase from the default input price. Similarly, we have analyzed the effects of identical percentage variations in the output price. We report the results for both price variations side by side.

Figure 6 depicts the optimal steady state phosphorus fertilization and the associated STP levels for different input (panel A) and output (panel B) prices. The rows under the graphs depict the associated numeric values. The leftward vertical axes depict the values of fertilization (kg ha^-1), whereas the STP values are given in the rightward vertical axes. The scales between the axes are dissimilar. They are, however, identical between the panels A and B. In the horizontal axis in panel A we have the values of input prices for which the steady states are calculated: 1.1 € kg^-1, 1.16 € kg^-1, 1.22 € kg^-1(the default value), 1.28 € kg^-1 and 1.34 € kg^-1. In panel B, the values for output prices are given in the horizontal axis: 0.1 € kg^-1, 0.105 € kg^-1, 0.11 € kg^-1 (the default value), 0.116 € kg^-1 and 0.12 € kg^-1.

Figure 6. Price variations and phosphorus.

A: input price and phosphorus

19.5

fert 20.8 20.6 20.4 20.3 20.1

STP 8.4 8.2 7.9 7.7 7.5

1.1 1.16 1.22 1.28 1.34

B: output price and phosphorus

19.5

fert 20.1 20.3 20.4 20.6 20.8

STP 7.5 7.7 7.9 8.2 8.4

0.1 0.11 0.11 0.12 0.12

In conformity with the analytical model, the graphs show that increasing the input price reduces the steady state fertilization and lowers the steady state STP. Increasing the output price has an opposite effect.

The effects of input and output price variations on optimal steady state allocations are approximately of the same order of magnitude. Varying the default input price from -10% to +10% lowers the steady state fertilization (20.4 kg ha^-1) by 3.3% and the STP value (7.9) by 10.8%. The increase associated with a similar variation in output price is 3.5% for fertilization and 12.1% for the STP.

Since the price variations affect the optimal use of phosphorus, they will affect the crop yields and profits as well. The steady state crop yield associated with the highest input price (1.34 €) was 0.7% lower than initially. For the lowest price (1.10€) it was 0.6% higher. The percentage changes associated with the output price variation were 0.7% for the lowest price (0.10 €) and 0.6% for the highest (0.12 €).

A change in the input price changes the private profits only slightly: a 10% increase in the input price decreases the steady state profit by 6.0%. A similar change in the output price, however, increases the profits by about 52.8%. The reason for this is that the fertilizer costs are only one part of the overall costs which are left intact as the fertilizer price is varied. On the other hand, the sales revenues are the only source of returns. Therefore, varying the output price changes the profits substantially.

Discount factor

In figure 7 we display the optimal steady state fertilization and STP levels for 5 different discount factors: 0.9, 0.925, 0.9, 0.975 and 1 (corresponding discount rates:

11.1%, 8.1%, 5.3%, 2.6% and 0%, respectively). Discount factor equal to unity means that the farmer does not discount the future profits at all. The leftward vertical axis depicts the level of annual phosphorus fertilization (kg ha^-1). The rightward vertical axis depicts the STP level associated with this optimal steady state fertilization. The horizontal axis denotes the values of the discount factor. The numeric values for the examined variables at the respective steady state optimum are under the horizontal axis.

Figure 7. The discount factor and phosphorus.

Discount factor and phosphorus

18.0 19.0 20.0 21.0 22.0 23.0 24.0

fertilisation (kg/ha)

5.0 7.0 9.0 11.0 13.0 15.0

STP (mg/l)

fert 18.4 19.4 20.4 21.7 23.5

STP 5.7 6.7 7.9 9.7 13.0

0.9 0.93 0.95 0.98 1

With a discount factor 0.9 the optimal steady state phosphorus fertilization is 18.4 kg ha^-1 which corresponds to an optimal steady state STP level of 5.7. With a default discount factor 0.95 the optimal fertilization is over 20 kilograms per hectare and the associated steady state STP level is 7.9. With a discount factor 1 the farmer considers only the edaphic dynamics and chooses an annual level of fertilization (23.5 kg) which corresponds to a significantly higher STP value: 13.0.

The results underline the importance of the role of the economic, dynamic decision making on the optimal choices. We discussed this already in the two-period example in section 3.1. Phosphorus fertilization can be considered mainly as an investment for

the future which – like all investment decisions – is affected by the discount factor used in the decision making process.

Again, the profits and crop yields are equally affected by the above variations, according to the new optima in phosphorus use. As we increase the discount factor, for instance, from 0.95 to 0.975, the steady state crop yield increases by 1.8%.

Similarly, the steady state profit per hectare increase by about 7.0%. That is, even though there are no changes in price ratios or in production technologies, the differences in the discount factor of the decision maker affect the steady state outcomes. Lower discount rate induces the farmer to make decisions that increase her steady state profits. This effect, be it surprisingly significant, is nevertheless logical.

Lower discount rate makes investments on soil more favourable. Investing in soil phosphorus status increases the steady state profits.

The effect of the transition function

Finally, let us examine how the above presented results would change if we were to use the alternative function to describe the development of the STP. Particularly, we want to study how the effect of the discount factor variation on optimal steady state allocations and outcomes will change when using the alternative transition function 47. To keep a given STP level unaltered, the transition function (47) requires higher annual fertilization than the transition function (45) (see also figure 5).

We first present side by side the steady state fertilization and STP levels sustained by the different transition functions. Panel A uses the (default) transition function (45) derived from Saarela et al (2004), and presents the privately optimal allocations for various discount factors. In panel B, we have a similar analysis using the transition function (47) by Ekholm et al (2005). In both panels the vertical axes on left denote the annual phosphorus fertilization (kg ha^-1). The vertical axes on right depict the STP levels at the various steady states. The horizontal axes depict the discount factors used while deriving the optimal steady states and the boxes under the horizontal axes depict the numeric values of the examined variables at the corresponding steady states.

Figure 8. The discount factor and phosphorus for different transition functions.

fert 18.4 19.4 20.4 21.7 23.5 STP 5.7 6.7 7.9 9.7 13.0

fert 24.7 26.1 27.5 29.1 31.1 STP 5.1 6.1 7.3 9.2 13.3 0.9 0.93 0.95 0.98 1

The comparison is interesting. The fertilization levels are approximately 32% - 35%

higher if we use the transition (47) than if we use (45). Surprisingly however, the STP levels are fairly close to each other in both cases. The steady state STP using (47) is about 10% lower at the lowest level of discount factor. With discount factor 1, the STP is about 2% higher than when using (45). At the default discount rate (discount factor 0.95), the steady state STP associated with (47) is about 8% lower.

Based on this comparison we might conclude that the optimal amounts of phosphorus fertilization are very sensitive towards the scientific modeling choices of edaphic processes. However, the optimal STP levels are not as strongly affected. They seem to be more sensitive on the economic behavior of the decision maker, in particular on the appreciation of future profits, captured by the discount factor. This has implications towards instrument design. This will be discussed in more detail in chapter 7.

Acknowledging the above results, the comparison between the steady state returns is fairly anticipated. The crop yields are very close to each other. Because of the differences in optimal phosphorus fertilization levels, the profits differ more than crop yields. With the discount factor 0.95, the profits per hectare when using the transition function (47) are about 13.2% lower than when using the transition function (45).