Economics of Individual Decision-Making: Buy Vs Lease Differences in the Adoption Of

Economics of Individual Decision-Making: Buy vs Lease Differences in the Adoption of Residential Solar

Dr. Varun Rai, The University of Texas at Austin – LBJ School of Public Affairs, 512-471-5057,

Ben Sigrin, The University of Texas at Austin, 651-983-6067,

Overview

We use a uniquely rich dataset from the rapidly growing residential photovoltaic (PV) sector to study individual decision-making. We focus on what financial metrics and information decision makers base their decisions on, and how their decision is affected by the structuring of rebates and business models. In particular, we study how the leasing and buying models affect individual choices. We find that in the early market we studied (Texas), a majority of PV adopters use payback period not net present value (NPV) as the decision-making financial criterion. Those who opt to buy PV systems display a systematic optimism in the value of solar (Figure 1). On the contrary, those who lease typically have a much tighter cash flow situation, which in addition to less uncertainty about technological performance when leasing, are the main reasons for them to lease. We also are able to calculate individual-level discount rate using a measure of implied net present value. Across a range of scenarios, buyers of PV systems have discount rates of 8-21% lower than the leasers (Table 1 & 2). Overall, our results suggest that the leasing business is able to address informational requirements of consumers more effectively than the buying model, and that the leasing model has also opened up the residential PV market to a new, and potentially a very large, consumer segment—those with a tight cash flow situation.

Figure 1: Comparison of payback period calculated by model to the period reported by consumer in our survey using baseline parameters assumptions: Electricity prices grow 2.6% annually; system lasts 25 years with a 0.5%/yr production loss; Buyers pay $0.7/W for inverter replacement and 0.25%/yr of system’s cost in maintanence; Consumer adopts any ‘solar’ plan offered by their current REP to receive outflow credits.

Data and Methodology

We use a novel data set which we have built by combining survey data on the behavioural, financial, and informational aspects of solar adoption withdata onthe consumers’.electricity consumption and rates as well as technicaldata of their PV systems. The sample population is 211 PV adopters in northen and central Texas who made their PV investment in 2010-11. The nature of the data allows very precise financial calculations at the individual level.

Using the above data we created a financial model which projects the revenues and costs associated with system ownership. The model calculates a number of standard financial metrics including NPV, internal rate of return (IRR), and Payback Period, which are compared to those the consumers reported using during their decision-making process. Finally, combining both the results from the financial model and survey we calculate the consumer’s discount rate based on a measure of implied NPV-- “I would not have installed the PV system if it had cost me [$1000 -$5000] more.”

Results

A majority of consumers reported using payback period, not NPV, as their decision-making financial criterion. This suggests that PV owners are not in the neoclassical rational-actor mould, which would require them to use NPV as the decision criterion. The payback period that consumers reported using in their decision-making process (reported) was compared with one generated by a financial model (modelled)that used detailed parameters (annual system production, system cost, consumer’s electricity rate, etc.) specific to each consumer. Across all scenarios, buyers were found to not only have higher payback periods than leasers, but to consistently underestimate the amount of time required to breakeven on their investment. In other words, buyers assumed favourable future conditions when assessing their investment; leasers were more realistic[1].

Buyersemployed a mean annual discount rate (DR) of 7%in the baseline scenario and leasers of 21% when assessing their investment in a residential PV system based on a series of questions that revealed their implied NPV. Discount rates are similar under a variety of scenarios with pessimistic or optimistic future assumptions, but are consistently significantly higher for leasers than buyers. Unlike Hausman (1979) we find no inverse correlation between income and implied DR.

Conclusions

Previous behavioral theory (Hausman, 1979) explains differences in investment behavior based on demographics, particularly income. In the context of the early residential PV market in Texas, this study finds no significant demographic difference between buyers and leasers, instead supposing that the two represent fundamentally different segments of the consumer market. Buyers imply a discount rate between 6-9% and therefore are more long-term and optimistic in their investment decisions (fig. 2) whereas leasers are realistic and cash-poor—which is reflected in the relatively higher (19 - 22%) discount rate they use.

References

Hausman, J. (1979). Individual Discount Rates and the Purchase and Utilization of Energy-Using Durables, Bell J. Econ. 10:1, pp. 33– 54.

Table 1. Mean Implied discount rate for buyers along income and scenarios with ±1σ.

Buyers / Implicit Annual Discount Rate
Annual Income / All Incomes / $0 – $85k / $85k – $150k / $150k+
N / 81 / 22 / 37 / 22
Scen 2: Pessimistic / 6% ± 6% / 6% ±5% / 6% ±8% / 7% ±6%
Scen 3: Baseline / 7% ± 5% / 7% ±4% / 6% ±6% / 7% ±6%
Scen 4: Optimistic / 13% ± 6% / 12% ±5% / 13% ±6% / 13% ±7%
Scen 5: V. Optimistic / 18% ± 7% / 17% ±5% / 18% ±7% / 17% ±8%

Table 2. Mean Implied discount rate for leasers along income and scenarios with ±1σ.

Leasers / Implicit Annual Discount Rate
Annual Income / All Incomes / $0 – $85k / $85k – $150k / $150k+
N / 81 / 22 / 37 / 22
Scen 2: Pessimistic / 20% ± 15% / 22% ±19% / 20% ±14% / 18% ±12%
Scen 3: Baseline / 21% ± 14% / 23% ±18% / 22% ±13% / 19% ±12%
Scen 4: Optimistic / 32% ± 17% / 33% ±22% / 35% ±15% / 30% ±14%
Scen 5: V. Optimistic / 35% ± 13% / 29% ±9% / 38% ±13% / 36% ±16%

[1] Mean difference between modeled and consumer payback period: Buyers = 7.1 yrs; Leasers = 1.1 years.