3 ms·
Sacramento municipal has a lot of customers in a small space. Running a grid in and around a major city is significantly different than covering an entire stat
by secabeen 1y ago
Sacramento municipal has a lot of customers in a small space. Running a grid in and around a major city is significantly different than covering an entire state with lots of mountains and forests from Oregon to San Luis Obispo.
- AlotOfReading 1y agoPG&E is expensive even when compared to the entire Western grid. I went and added up the revenues for all the major utilities in Arizona, Colorado, Nevada, New Mexico, Oregon, and Utah. That's APS, NV Energy, Pacificorp, PSCo, PSCNM, and SRP. All of their 2024 revenues combined are ~$22B vs PG&E's $24.4B, to cover an area 3x larger with just over half the population and even drier climates. All of those states have lower average rates as well. PG&E is uniquely expensive for no legitimate reason that I've ever been able to discern.
- wahern 1y agoThe jump in PG&E prices over the past 10 years is mostly to cover the cost of wildfire mitigation. LA, Palo Alto, Santa Clara, and the Imperial Valley have their own local utilities that charge customers 1/2 to 1/4 of what PG&E charges.[1] But PG&E covers the entire northern half of the state. Bay Area PG&E customers are subsidizing all those small communities nestled in the forests. Southern California Edison and San Diego have a mix of urban and forested rural closer to PG&E, with significant wildfire mitigation costs of their own, and their rates reflect that--2/3 to 3/4 of PG&E prices). Despite approving the cost increases, state regulators are now complaining that PG&E is spending too much on wildfire mitigation. For example, they're telling PG&E to shift from burying transmission lines to using covered conduction (i.e. plastic shielding around overhead transmission lines). But PG&E is pushing back because they're on the hook for the wildfire civil claims and argue the less costly mitigations are too risky. [1] See https://www.reddit.com/r/bayarea/comments/196kjtu/pge_rates_vs_other_ca_utilities/ https://www.reddit.com/r/bayarea/comments/196kjtu/pge_rates_...
- AlotOfReading 1y agoAll of that also applies to pacificorp, which covers the Jefferson area of California. I'm actually double counting a bit in PG&E's favor as a result.
- wahern 1y agoThe wildfire hazards aren't equivalent, nor the political pressures regarding those hazards. But it seems Pacific Power isn't too far behind: https://oregoncub.org/news/blog/pacific-power-asks-for-21.6-rate-increase-for-2025/2979/ https://oregoncub.org/news/blog/pacific-power-asks-for-21.6-...
- glompers 1y agoIt seems to an outsider like California involves exceptional biodiversity, and exceptional habitat protection laws, coupled with the electorate's vocal idealism about the possibility of preventing preventable animal suffering, plus the distaste of many voters for a schedule of very frequent controlled burns in all parts of the state in order to thin the constant buildup of combustible underbrush. These leave PG&E unable to avoid being liable for interactions of transmission lines with the wildfire kindling they have to route through. Even though California may not be more mountainous than those other Western states, its development pattern may have more wildland-subdivision interface overall, which adds up.
- p_j_w 1y agoIf you actually meant revenue when you said revenue, then it shouldn't be surprising that PG&E has higher revenue compared to companies with half the population. Revenue is merely income, no costs will be deducted from that number. So, if PG&E has the same revenue but twice the customer base, that means they're extracting on average, half the money per customer as those other companies, which doesn't make them sound more expensive! Did you mean profit?
- AlotOfReading 1y agoUtilities put most of their revenue back into costs and I'm not interested in the relative profitability, so I really do mean revenue. I'm assuming that users correlate to revenue, and the primary cost driver is service area, which means the efficiency is the ratio between area and revenue. I'm also making the simplifying assumption that the major utilities service their entire states and that the area being discussed is large / demographically similar enough that a lot of the other differences average out. These aren't true (e.g. PG&E only services part of California), but I'm hoping it's sufficient for a sanity check. The ratio of revenue to area served here is radically different though and the other states are significantly better. We can make a prediction to double check this that people in those states pay lower rates on average, which is also true. If you want to go and add up the customers directly, PG&E serves 5.5M. The others serve around 8.3M people collectively from some quick searches. I don't have info on actual service area, nor do I really want to spend months writing an industry report on it.