--- id: VOLT-HOME-WP-084 title: "How much fixed-price hedging should a flexible household retain?" slug: how-much-fixed-price-hedging-should-a-flexible-household-retain description: "A transparent volatility-ratio proxy for splitting standardized household load between fixed and dynamic wholesale exposure." published: 2026-08-30 cluster: "Tariff economics and consumer safeguards" status: measured evidence: /research-data/home-papers/how-much-fixed-price-hedging-should-a-flexible-household-retain.json figure: /research-media/home-papers/how-much-fixed-price-hedging-should-a-flexible-household-retain.webp figure_alt: "Chart for How much fixed-price hedging should a flexible household retain?: variance-minimizing fixed-price hedge proxy, shown as dynamic, fixed hedge." source_ids: - dynamic-tariff-viability - acer-retail-2025 - ec-retail-flexibility-2026 - energy-policy-flexibility - iea-demand-flexibility peer_reviewed: false --- ## Abstract This working paper reports a simple fixed-price hedge proxy for a standardized synthetic household exposed to daily mean wholesale electricity prices. The proxy divides the population standard deviation of observed mean prices by the sum of that standard deviation and the absolute sample mean, then expresses the ratio as a percentage of load. Across 93,161 observations, the result is 74.83134049898806%. The percentage is a transparent volatility ratio, not the solution to a portfolio optimization, not a recommended retail-contract share, and not financial advice. The family scenario declares 4,000 kWh annual use and 35% flexible energy, but neither quantity enters the proxy formula. No fixed-price premium, supplier credit risk, contract duration, consumption-price covariance, taxes, network charges, VAT, device efficiency, or customer telemetry is modeled. The result therefore answers a narrow measurement question: how large is one rule-based fixed-share proxy under the frozen wholesale-price distribution? It does not establish how much any household should hedge. ## Plain-language answer The frozen formula produces a fixed-price share of 74.8313% of load. That does not mean households should fix 74.8313% of their electricity. It means the selected ratio maps the observed level and volatility of daily mean wholesale prices to that percentage. The formula rises when wholesale-price volatility is large relative to the absolute average price and falls when volatility is small relative to that average. This is intuitive as a risk indicator, but it is not enough to choose a contract. A real fixed price includes a supplier’s hedge cost, margin, risk premium, credit exposure, and contract terms. A household’s risk also depends on when it consumes. If its load is highest on expensive days, exposure differs from that of the flat standardized profile. If flexible devices can avoid high-price intervals, the desired fixed share may differ again. No euro saving is reported in this paper. Any monetary comparisons discussed are synthetic wholesale scenarios only, never customer savings or retail bills. The evidence itself calls the result a risk proxy rather than an optimized retail contract. The practical answer is therefore: 74.8313% is the output of the declared heuristic, not a personal recommendation. ## Research question The broad question is how much fixed-price protection a flexible household should retain while preserving some benefit from dynamic prices. The implemented research question is narrower: what fixed-load percentage results from a preregistered, transparent function of the pooled mean and standard deviation of daily wholesale prices? The study does not estimate expected utility, risk aversion, affordability, or an optimal hedge. It does not compare actual fixed and dynamic offers. “Retain” is used in a scenario sense, not as an instruction to renew, buy, or sell a financial product. This boundary matters because a statistically convenient ratio cannot replace a household’s tariff, liquidity, and consumption constraints. ## Data and provenance The result comes from a frozen, public-safe production snapshot extracted with SELECT-only access. No customer telemetry is included. The tariff family registers `day_ahead_prices`, `forecast_accuracy`, `bess_index_daily`, `zone_load`, and `zone_holidays`; the primary proxy uses the collection of daily mean day-ahead prices. The matching evidence reports 93,161 observations. The daily-price window is 2021-01-01 through 2026-08-29. Detailed intervals in the corpus span 2025-10-01 through 2026-08-29, and the long-history boundary spans 2015-01-01 through 2026-08-29. The publication cutoff is 2026-08-30T00:00:00Z. The additional contracts and windows form the registered evidence surface, but they should not be represented as inputs to the simple volatility ratio when they are not in its formula. The source snapshot has SHA-256 `7e97489fc8528c8cc8c38830e05b48d949ce1f67b98575dff26f5d7c321e4c67`. The preregistration hash is `adb36bf6b447af9f96339249b8becaefc20422499cca1977242866347a97bd4b`; the paper-registry hash is `7bcb91d7476d0a69fe9fa75a5c7782f8117e0153f82f9112b7e1d307d3943717`; the source-registry hash is `07949550ac443ff673fda5c0209b99f137544f3ffecf6775f109bb9d09663bd6`; and the analysis-code hash is `57c57de79cdab2b5b6d6c54c485cb5162598c5ba0b0bfe995da40d75e6c52ba9`. The evidence-manifest hashes are `366a3d93f435cfaa6f1d5e771482bca959f976182896adb6240af295f51a0467` for the JSON and `d21c2c99eeb593b5de99f1e2912d4964e520ffcc89c03541f3c75ea9d3006592` for the WebP figure. Extraction used a read-only transaction with a 180-second statement timeout. The registered references provide external context, while the frozen Volt evidence is the sole basis for the reported proxy. ## Method Let `m` be the arithmetic mean of all included daily mean wholesale prices and let `s` be their population standard deviation. The fixed-price hedge proxy is: `100 × s ÷ (absolute(m) + s)`. The implementation clips the result to the range from 0% to 100%, although the ratio is already non-negative when its inputs are finite. The figure divides the scenario into a “fixed hedge” component and the remaining “dynamic” component. The phrase “variance-minimizing” in the registered metric name should not be overread. A conventional minimum-variance hedge normally requires a return or cost covariance between the hedged exposure and the hedge instrument, as well as a defined objective and constraints. Here there is no separate hedge-instrument price series and no covariance optimization. The function uses only the mean and standard deviation of the exposed wholesale-price series. We therefore describe it consistently as a volatility-ratio proxy. The family assumptions declare 4,000 kWh annual synthetic load, 35% flexible energy, wholesale pass-through, and no customer telemetry. The first two quantities do not enter the formula. The proxy applies to a percentage of abstract load, without distinguishing flexible and inflexible consumption. There is no actual fixed-price contract, maturity, premium, cancellation term, or counterparty. No household asset is simulated. Battery efficiency, degradation, state of charge, and inverter power are absent. EV charger efficiency, power, arrival, and departure are absent. Heat-pump COP, comfort, and thermal storage are absent. Taxes, network charges, VAT, supplier margin, and device losses are excluded. Any euro-denominated extension would be a declared synthetic wholesale scenario, not a retail result. ## Results The frozen primary result is 74.83134049898806% of load. Rounded to two decimals, the proxy assigns 74.83% to fixed-price coverage. The sample size is 93,161 daily-price observations. This percentage is evidence about the chosen formula under the chosen pooled wholesale-price distribution. It is not evidence that 74.83% minimizes a household’s bill variance. It is not evidence that the remaining dynamic share will produce savings, and it does not estimate the price charged for fixed protection. The figure divides the proxy into exactly 25.168659501011945% dynamic exposure and 74.83134049898806% fixed hedge. The regenerated evidence reports `bootstrap_95_interval: null` and `interval_method: "not reported for this estimand"`, so it supplies no uncertainty bounds for the fixed-share proxy. The key finding is consequently modest: the frozen heuristic places substantially more of the abstract load on the fixed side than the dynamic side. Whether that is desirable cannot be inferred without contract prices, household load covariance, and a declared risk objective. ## Robustness and placebo checks A constant-price placebo sets every daily mean price to the same positive value. The standard deviation becomes zero, and the fixed-share proxy becomes 0%. This reveals a feature of the rule: absent volatility, it sees no statistical reason for fixed protection even though a household might still value budget certainty. A zero-mean, volatile-price stress drives the ratio toward 100% because the denominator approaches the standard deviation. Negative prices and a mean near zero can therefore produce a very high fixed proxy even when average wholesale cost is low. The use of `absolute(m)` prevents a negative mean from creating a negative denominator, but it does not solve the economic interpretation problem. A regime robustness check would compute the proxy separately by zone, year, and crisis versus recent period. Pooling can combine markets with different currencies, price caps, and volatility structures unless the source corpus has already harmonized the relevant records. This paper reports the registered pooled result and does not claim stability across strata. A stronger comparator would minimize the variance of synthetic total cost over a grid of fixed shares while adding a declared fixed-price premium and preserving the household’s load path. It should be evaluated out of sample and compared with simple shares such as 0%, 50%, and 100%. That optimization was not performed. The family protocol calls for blocked day or week bootstrap intervals and Holm correction for inferential claims. This paper makes a descriptive report of a deterministic statistic. It does not infer an optimal policy from the observation count. ## Limitations The result is highly dependent on the proxy definition. Another monotone transformation of mean and volatility would produce another percentage. The label “variance-minimizing” is stronger than the implemented calculation supports because no hedge covariance or objective optimization appears in the method. The input is a pooled wholesale-price distribution rather than the cost series of a household. Consumption quantity and timing do not enter. The family’s 4,000 kWh load and 35% flexible share are contextual assumptions, not formula inputs. As a result, flexibility is not actually valued, and the study cannot determine how controllable devices change the preferred fixed share. Retail economics are absent. Fixed-price contracts usually embed premiums and terms that may vary by provider and period. Dynamic contracts can include markups, caps, floors, and subscription fees. Taxes, VAT, supplier margin, and network charges are excluded. No counterparty or default exposure is considered. There are no physical asset assumptions because no asset dispatch occurs. Efficiency and degradation are therefore not zero; they are unmodeled. The distinction prevents readers from interpreting the proxy as a battery or EV recommendation. Historical mean and volatility can change. The result may be sensitive to crisis periods, negative prices, zone composition, and the publication cutoff. The evidence reports no uncertainty interval for the percentage. Finally, preferences differ: variance minimization, expected cost, tail protection, and budget smoothness are separate objectives. ## Practical implication Treat 74.8313% as a diagnostic starting point for a scenario grid, not as a contract instruction. A decision-grade analysis would calculate total synthetic retail cost for several fixed shares using the household’s actual or standardized load path, local tariff components, fixed-price premium, dynamic markup, and explicit risk measure. Flexible appliances should then be optimized under physical constraints for each share. The chosen policy may favor more or less fixed exposure depending on whether flexibility reduces expensive-day consumption. Any euro result from that extension must remain labeled a synthetic wholesale or retail scenario until validated against actual bills; this paper supplies neither. ## Reproducibility Use the evidence URL in the frontmatter to verify the ID, title, slug, assumptions, status, publication cutoff, source contracts, all five provenance hashes, and the evidence-manifest hashes for the JSON and figure. Confirm that the extraction is marked read-only and that the snapshot hash matches the corpus manifest. From the frozen daily-price records, collect each `mean_price`. Compute the arithmetic mean and population standard deviation exactly as in the content-addressed analysis code. Evaluate `100 × standard deviation / (absolute(mean) + standard deviation)` and clip to the interval from 0 to 100. The result should reproduce 74.83134049898806%. Do not substitute sample standard deviation without declaring the change. Do not add a fixed-price premium, weight by household load, split by zone, or optimize a covariance hedge and call it a reproduction. Each would be a valuable new study with a new evidence record. ## Disclosure Analysis and drafting were model-assisted. This public working paper is not peer reviewed and discloses its formula, assumptions, evidence hashes, and limitations. It is not financial, investment, purchasing, tariff-selection, or trading advice. No measured customer saving or retail bill is reported. Any monetary context is a declared synthetic wholesale scenario. Volt has no live traders or live capital; C0R is the only paper strategy. This household-risk proxy is not trading research or a trading recommendation. ## References 1. Advances in Applied Energy, [Assessing the conditions for economic viability of dynamic electricity retail tariffs for households](https://doi.org/10.1016/j.adapen.2024.100174). 2. ACER and CEER, [Rewarding flexibility: How retail contract choice can help unlock consumer flexibility](https://www.ceer.eu/wp-content/uploads/2025/11/ACER-CEER-2025-Retail-monitoring.pdf). 3. European Commission, [Communication from the Commission on the Citizens Energy Package](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026DC0115). 4. Energy Policy, [Welfare redistribution through flexibility — Who pays?](https://doi.org/10.1016/j.enpol.2025.114684). 5. International Energy Agency, [Scaling Up Demand Flexibility](https://www.iea.org/reports/scaling-up-demand-flexibility).