Published to LuthINformed, Issue 7 (July 20, 2016)
In this issue, we offer a look at energy reporting best practices. As the old adage goes, you cannot manage what you cannot measure. We hope to provide our readers with guidance and actionable information that will be both valuable and useful. As always, your feedback is welcome…
Best Practices in Anomaly Reporting…
The process of looking for energy billing anomalies can increase efficiency and lower costs. Energy usage is generally consistent by season (summer/winter), annual periods (last year/this year) or monthly periods (last month/this month). An anomaly does not necessarily mean that a bill is incorrect. Other factors such as weather, operational changes and fuel switching may create anomalies. In a normal weather scenario we would recommend starting with a 25% variance. The challenge is to find a variance threshold that is small enough to detect meaningful errors without generating excess false positives. Ideally, a variance factor would be consistent across an entire portfolio of accounts. However, this is not always practical. For example, a sewerage water treatment plant’s energy usage may be related to rainfall; or, it may have seasonal characteristics unlike office buildings. The variance factors for these accounts may be different for those reasons. Also, for high dollar accounts, which are generally fewer in number, a lower variance factor may be employed since missing one error can have a significant cost impact.
Unit Cost Variance Analysis…
A common best practice in auditing is to conduct a Unit Cost Variance Analysis. With increased emphasis on benchmarking in the energy business, the concept of normalization comes into play when using unit analysis. We have found it effective to calculate unit costs ($/kWh, $/DTh, or $/Mlb) and then compare them to the median of some historical period. Ideally, use two to three years of historical data for the analysis – but a twelve month period will suffice if longer periods are not available. Remember that cost components usually vary with seasons; for example, electric delivery rates are significantly higher in the summer because of an additional summer demand charge in the tariff. This may require employing filters to ensure an “apples to apples” comparison. We have found that the median is a better comparison than the average, since averages can be impacted by changes in usage pattern over time. Also, averages will be influenced by the anomaly itself. Again, where average is all you have, it is still an adequate comparison tool. Other types of normalization are based on degree days, facility area, or some other factor such as units processed.
Billing Determinants Analysis…
Billing Determinants Analysis is used to identify unexpected occurrences such as identical demand (see above) or usage across consecutive time periods. We take the position that there are no coincidences in utility billing! Utilities often estimate bills when actual data is unavailable. If a bill is calculated incorrectly, errors will not be identified by analyzing usage anomalies only. So when we see two consecutive bills with identical data, we consider that an anomaly – open for investigation. In a similar manner, bills that are revised more than once a year – or for more than a single billing period – should be scrutinized. We have found that when either of these two conditions occur, we are dealing with a form of estimated billing, meter problems or other systemic issues. Past experience has shown that utilities do not always code or label a bill as estimated.
Load Factor Report…
Identifies instances where load factors exceed 100%. The load factor is a ratio based on the peak demand to metered electric usage; it cannot be greater than 100%. A 100% load factor would indicate that the account used the peak demand during every hour of the billing period, therefore, a load factor greater than 1 is impossible. Load factor should be consistent with past usage. Incorrect meter readings, defective metering, or improperly estimated bills can be identified using load factor analysis that compares changes over the recent and historical billing record of the account.
Energy Change Report…
Shows total billed energy usage by facility, in millions of BTUs, for the month and the fiscal year-to-date. It also shows the year-over-year percent change in energy consumption adjusted for weather. An effective tool in identifying anomalies across facilities and campuses since a more global, not per-meter, perspective can help identify anomalies which are then analyzed by drilling down to the individual account level.
Natural Gas Outlook…
The latest inventory report (above) shows a net increase of 64 Bcf vs. the previous week. Stocks were 507 Bcf higher than this time last year and 586 Bcf above the five-year average. Immediately following the release of the report, the futures curve rose slightly despite a higher than anticipated build. Over the past two weeks, strip prices are down slightly. Current prices are listed below:
12 month strip = $3.013
24 month strip = $3.040
Cal Year 2017 = $3.118
Cal Year 2018 = $3.001
(all prices NYMEX only; A/O 7/19/16)
For an in-depth discussion on markets, purchasing strategies and other topics, call us here at Luthin Associates. We offer our clients regular market updates and our Energy Procurement Group is staffed with certified experts on energy market conditions.
Although care has been taken to ensure the accuracy, completeness and reliability of the information provided, Luthin Associates, Inc. (Luthin) assumes no responsibility therefore. The user of the information agrees that the information is subject to change without notice. Luthin assumes no responsibility for the consequences of use of such information, nor for any infringement of third party intellectual property rights which may result from its use. In no event shall Luthin be liable for any direct, indirect, special or incidental damage resulting from, arising out of or in connection with the use of the information.