Time-of-Use Battery Arbitrage Under NEM 3.0: Does It Actually Pay?

By Nacho Iniguez ✦ Updated June 10, 2026

If your utility charges you more for power at 7 pm than at 2 am, a battery can buy low and sell high inside your own house. That is time-of-use (TOU) arbitrage. It sounds like free money, and in some places it almost is. In others, after you account for conversion losses and wear, the spread barely covers the cost of running the battery. This guide walks through how it works, the actual math, and where it stops being worth it.

What TOU arbitrage really means

TOU rate plans price electricity by the hour. A typical California evening peak (often 4 to 9 pm) costs far more than overnight off-peak hours. Arbitrage means charging the battery when power is cheap and discharging it when power is expensive, so you avoid buying the expensive kWh from the grid.

There are two ways to fill the battery:

• From the grid overnight, at the off-peak rate. This is pure arbitrage and works even with no solar.
• From your own solar at midday, then holding it for the evening. This is the dominant strategy in California right now, and it is really a NEM 3.0 story.

The mechanic is the same either way: energy goes in at a low effective cost and comes out displacing a high-priced kWh.

Why NEM 3.0 changed the math

Under the old net-metering rules, a kWh you exported at noon was credited at nearly the full retail rate, so a battery was mostly about backup. Under NEM 3.0 (the Net Billing Tariff), exported solar is paid at low avoided-cost rates, often a few cents per kWh, while the power you pull from the grid at 7 pm can run $0.40 to $0.59 in summer peak.

That gap is the whole game. Storing 1 kWh of midday solar and using it at peak can be worth roughly ten times more than exporting that same kWh. The battery is not selling to the grid; it is letting you keep your own cheap energy and spend it when grid power is most expensive. For the bigger picture on whether a system pencils out at all, see our is solar worth it in 2026 guide.

The arbitrage math, kWh by kWh

The per-cycle value of arbitrage is close to:

savings ≈ spread × kWh discharged × round-trip efficiency

The spread is the difference between the price you avoid (peak) and the price you paid to charge (off-peak, or the export rate you gave up). Round-trip efficiency is how much energy you get back out versus what you put in. Modern AC-coupled home batteries land around 89 to 90 percent round-trip per the manufacturers, with DC-coupled paths a bit higher (Tesla and Enphase publish figures in the high 80s to mid 90s depending on configuration).

Worked example, grid-charged arbitrage with a $0.30 spread:

• Discharge 10 kWh at peak, avoiding 10 × $0.55 = $5.50 of grid power.
• You spent 10 / 0.90 ≈ 11.1 kWh charging at $0.10 off-peak = about $1.11.
• Net for the cycle: roughly $4.39.

Solar-charged is even better, because your charging “cost” is the export credit you gave up, often just $0.05 to $0.08 per kWh. To run your own numbers with your real rates and battery size, use the TOU arbitrage calculator, and pair it with the battery sizing tool so the pack is big enough to cover the full peak window.

Where the spread makes or breaks it

The single biggest variable is your utility’s spread. From 2026 rate data, the picture is uneven:

• SDG&E: summer TOU spreads around $0.37 per kWh on some plans. Strong arbitrage case.
• SMUD: roughly $0.22 per kWh peak to off-peak. Solid.
• SCE: often $0.14 to $0.22 per kWh, depending on plan. Meaningful but plan-sensitive.
• PG&E: after recent restructuring, some plans show only $0.10 to $0.13 per kWh.

At a $0.30 spread, arbitrage clears losses easily. At $0.10, after you subtract round-trip losses and a slice for degradation, the margin gets thin enough that pure grid arbitrage may not justify cycling the battery hard. Picking the right rate plan can matter as much as picking the right battery.

The limits: efficiency, degradation, and thin spreads

Three things quietly eat into the gross number:

• Round-trip losses. At 90 percent efficiency you lose about 10 percent of every kWh you cycle. On a thin spread, that loss can swallow most of the gross margin.
• Degradation. Cycling adds wear. Quality LFP packs lose roughly 2 to 3 percent of capacity per year in normal use and are typically warrantied to retain about 70 percent capacity after their rated throughput (often 8 years or several thousand cycles). Cycling daily for arbitrage uses up that throughput faster than backup-only use.
• The spread itself. Without a wide, reliable peak/off-peak gap, the math simply does not clear. Arbitrage rewards big spreads; it punishes flat rate structures.

A practical rule: if the spread comfortably exceeds your losses plus a small per-kWh allowance for wear, daily cycling pays. If it does not, lean on the battery for backup and bill-smoothing rather than chasing arbitrage.

Bottom line

TOU arbitrage is real money under NEM 3.0, but it is not automatic. The value comes down to spread times kWh times efficiency, and the spread is set by your utility and rate plan. In high-spread territory the case is strong; in low-spread territory the losses and wear can erase it. Run your own rates through the TOU arbitrage calculator, check the full payback in the solar-and-battery ROI calculator, and if backup ranks high on your list, compare hardware in our best home backup battery roundup. More buyer guides live in /guides/ and /reviews/.