Study this credential by practicing decisions, not definitions. For each topic, ask what a decision-maker must choose, which numbers settle the choice, and who holds the risk. Work scenarios in writing, compare financing structures side by side, and check your pro-forma math against plausible regional production ranges.
Turning engineering inputs into executive decisions
Treat every case prompt as a request for a decision: identify what the decision-maker must choose, which two or three figures drive the choice, and what risk shifts with each option before you write anything.
A design question asks whether a roof can hold a certain array; a business question asks whether the owner should sign, buy, or walk away. When you read a case, mark every technical fact and label what it feeds: irradiance data feeds the revenue estimate, roof age feeds the terminal-value and re-roofing question, and utility tariff structure feeds the avoided-cost calculation. This labeling habit converts a wall of facts into a short list of decision variables.
Practice the conversion explicitly. Take any paragraph of case facts and write one sentence beginning 'The decision is whether to...' followed by the two options on the table. Then list the numbers that differ between those options only. Anything that does not differ is context, not analysis. In timed practice this discipline matters because a complete answer usually hinges on three or four numbers, and spending your limited writing time on neutral background produces an answer that describes the situation without resolving it.
LCOE, NPV, IRR, and payback answer different questions
Levelized cost of energy compares energy costs across options with different lifetimes and output; NPV and IRR measure value to capital; simple payback measures only how fast cash returns. Name the question before picking the metric.
LCOE expresses total lifetime cost divided by total lifetime production, in currency per kilowatt-hour, which makes it the right tool for comparing an owned system against a purchased-power alternative over unequal time horizons. Simple payback ignores everything after the break-even year and ignores the time value of money, so it answers a liquidity question, not a value question. Using payback to judge a long-lived asset systematically undervalues years of production that arrive late but still matter.
NPV discounts future cash flows to today's value using a discount rate that represents the capital's opportunity cost, while IRR is the discount rate at which NPV reaches zero. The practical difference: NPV tells you how much value a project adds at a known cost of capital, and IRR tells you the return the project itself earns, which is useful when comparing projects of different sizes. In written answers, state which metric you are using and why, because a case asking 'is this a good use of the company's capital' is an NPV question even if the facts are presented as payback figures.
Comparing the main financing structures side by side
Cash purchase, loan, lease, and power purchase agreement differ in who pays upfront, who claims incentives, how the energy price behaves over time, and which risks the host keeps. Build this table from memory.
In a cash purchase the host pays everything upfront, owns the equipment, claims available incentives where eligible, and carries performance and maintenance risk. A loan shifts the upfront burden to a lender while the host still owns the system, so most of the ownership profile remains, with repayment obligations added. These two keep ownership with the host; the next two do not.
Under a lease the host pays a fixed periodic amount for the use of the equipment, and under a power purchase agreement the host pays per kilowatt-hour actually produced, which moves production risk to the system owner. In both third-party structures, the arrangement of incentives depends on who owns the equipment and on the tax rules of the specific jurisdiction, so a case answer should say 'the equipment owner is positioned to claim production- or investment-based incentives, subject to local tax rules' rather than asserting a specific benefit. The table below summarizes the structure you should be able to reproduce in an exam answer.
| Structure | Upfront capital by host | Who owns equipment | Energy cost profile | Key host-side consideration |
|---|---|---|---|---|
| Cash purchase | Full system cost | Host | Declining to near zero after payback | Performance and maintenance risk |
| Loan | Small or none | Host | Fixed repayments, then low cost | Debt service versus savings |
| Lease | Little or none | Third party | Fixed periodic payment, possibly escalating | Total lease cost over term |
| Power purchase agreement | None | Third party | Price per kWh, often with escalator | Escalator versus utility rate growth |
Worked scenario A: the PPA escalator trap
The classic error is comparing a PPA's year-one rate to the current utility rate. Escalators compound, so the honest comparison uses the levelized PPA rate against the owner's levelized cost of ownership.
Simplified scenario: a commercial host consumes 700,000 kWh per year. A 20-year PPA offers energy at $0.11 per kWh in year one with a 2.5 percent annual escalator; the utility rate today is $0.13 per kWh. The tempting conclusion is that the PPA saves money from day one. But summing the escalating PPA payments over 20 years gives a levelized PPA rate of roughly $0.14 per kWh, higher than today's utility rate, so the deal only wins if utility rates grow faster than the escalator. The better decision process compares three levelized numbers: PPA, utility forecast, and ownership.
Now add the ownership option: the same array purchased for $1,050,000 before incentives, producing 700,000 kWh per year, implies an unsubsidized cost of about $0.075 per kWh over 20 years before applying any local incentives and before deducting operating costs, which a careful answer must also include. The point is not that ownership always wins; it is that the escalator changes which comparison is honest. In your written answer, show the levelized PPA calculation explicitly, state the utility-rate growth assumption you are testing against, and note that the conclusion flips at different escalator and rate-growth values. That sensitivity statement is what distinguishes an executive answer from an arithmetic one.
Worked scenario B: stress-testing the production line of a pro-forma
A pro-forma's revenue is only as credible as its production estimate. Check specific yield in kWh per installed kW, confirm whether figures are DC- or AC-based, and apply degradation rather than using year-one output for every year.
Simplified scenario: a pro-forma shows 1,200,000 kWh in year one from a system described as '1 MW.' The plausible mistake is accepting that number and building revenue on it. First ambiguity: is 1 MW the DC array size or the AC inverter capacity? If 1.25 MW-dc feeds a 1 MW-ac inverter, the specific yield differs by 25 percent depending on which base you use. Dividing 1,200,000 kWh by 8,760 hours gives a capacity factor of about 13.7 percent on an AC basis, which is plausible for a sunny region with a well-oriented fixed array but high for cloudy climates; the same output on a 1.25 MW-dc basis is a lower, more conservative yield figure.
The better decision path recomputes the estimate independently: convert the stated capacity to specific yield in kWh per kW, compare that yield against ranges typical for the region and mounting type, then apply a degradation assumption so that year ten and year twenty revenues are lower than year one. In the example, a 0.5 percent annual degradation reduces average annual output over 20 years by roughly five percent relative to year one, which flows straight through revenue. Your written answer should state the DC/AC basis you assumed, cite the yield range you benchmarked against, and note that financing terms often depend on production, so an inflated estimate changes both revenue and covenant headroom.
Naming risk allocation in contract and case answers
Executive answers assign each major risk to a party: EPC price and schedule, permitting and interconnection approval, long-term production and O&M performance, and off-taker credit. Unassigned risk reads as incomplete analysis.
An EPC contract typically places construction price and completion-date risk on the contractor, which is why fixed-price turnkey terms matter to a financer; a case where the developer carries construction overruns themselves signals a weaker structure. Interconnection approval usually sits outside anyone's control and is often the schedule risk no contract fully transfers, so a strong answer identifies it as a gating milestone rather than pretending it is managed. Reading a contract excerpt in a case means asking, clause by clause, who pays if this event occurs.
Operating-phase risks split differently: under a PPA, production risk generally sits with the system owner, who responds with O&M agreements and availability guarantees, while the host's main exposure becomes the off-take price and escalator over time. Credit risk runs in both directions: the owner carries the host's payment obligation, and the host carries the owner's ability to maintain the system for twenty years. In a written answer, a short risk table with the risk, the party bearing it, and the contractual mechanism is an efficient, high-signal format, and it forces you to notice when a case leaves a risk, such as inverter replacement after warranty expiry, assigned to nobody.
Practice exercise, self-check rubric, and preparation sequence
Build one mini pro-forma from a real utility bill and a hypothetical quote, then score yourself against a rubric covering metrics, production assumptions, financing choice, and risk assignment before moving to full case write-ups.
Exercise: take a real or realistic annual utility bill, choose a hypothetical commercial array size, and produce a one-page analysis containing four elements: a year-one production estimate with your stated specific-yield assumption, a levelized cost of ownership, a levelized PPA rate at a chosen escalator, and a three-row risk table. Then write a five-sentence recommendation naming one decision, its two or three driving numbers, and the condition under which you would reverse it. Expected observations when you review your own work: the recommendation usually rests on two numbers you can name in one sentence, and your first draft almost always leaves one risk unassigned.
Self-check rubric, scored as learning milestones rather than pass predictions: two points for using the metric that matches the question asked, two for a production estimate with an explicit DC/AC basis and degradation, two for a levelized comparison rather than year-one rates, one for naming every major risk's bearer, and one for a stated sensitivity or reversal condition, giving a self-check target of eight of ten before you attempt full case scenarios. An adaptable sequence: first week, rebuild the metric definitions and the financing-structure table from memory; second week, run the exercise above twice with different tariffs; third week, write two full timed case answers and grade them with the same rubric; final stretch, redo your weakest scenario type until the rubric score holds.
Readiness checks before exam day: you can reproduce the financing comparison table without notes, explain in two sentences why payback alone misjudges a long-lived asset, compute a levelized PPA rate with an escalator in under five minutes, and state which party holds construction, interconnection, production, and credit risk in each structure. Administrative details such as scheduling and eligibility belong to the issuer; Solar Energy International's site, linked in the sources, is the appropriate reference for those logistics.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.