[Discussion] TRON Quarterly Transaction Fees Review — Q2 2026

[longgucun-yk]

Summary

This is a Q2 2026 follow-up tracking the energy fee reduction (unit price 210 → 100 sun, effective 2025-08-29), continuing #834 (Q1 2026) and extending each framework through 2026-06-30. Every framework below was checked against #834's own numbers before extending (see Data & method).

Headline: the two dynamics #834 identified both continued into Q2, but pulled apart: the structural-inflation side has clearly deepened, while the activation side narrowed. Direct-burn share of energy fell to ~10.0% by Jun 2026 (from 11.7% at #834), staking now supplies ~90% of energy, and the annualized net supply change rose to +0.44% in Jun 2026 — the highest monthly reading in the window. On the activation side the $1–10 micro-payment band strengthened further (+65%, up from #834's +56%), but the gain did not broaden — the $10–1000 core band held roughly flat (~+8%), and a sub-$1 dust tier reversed to −17%. Energy demand itself has plateaued in a ~190–203B/day band.

Quarterly cost table (pre-cut baseline → 2026-Q2)

Period	Avg TRX Price	Energy Unit Price (sun)	USDT Transfer Fee (Burn)	SunSwap V3 Fee
2025-Q2 (pre-cut)	$0.26	210	$3.52	$12.68
2025-Q3 (pre-cut)	$0.33	210	$4.39	$15.80
2025-Q3 (post-cut)	$0.34	100	$2.18	$7.85
2025-Q4	$0.30	100	$1.90	$6.85
2026-Q1 (Jan/Feb)	$0.29	100	$1.87	$6.73
2026-Q2	$0.33	100	$2.14	$7.72

The 210 → 100 sun cut roughly halved on-chain fees and they have stayed there: a USDT transfer went from $3.52–4.39 (pre-cut) to ~$1.87–2.18, and a SunSwap V3 swap from $12.68–15.80 to ~$6.73–7.85. Fees ticked up modestly in Q2 as TRX recovered ($0.29 → $0.33) but remain far below the pre-cut level. Method: reference energy × unit price (sun) × avg TRX price (USD/TRX), scaled by 1e-6 — where reference energy is 64,285 for a USDT transfer to an existing-balance address (a first-time / zero-balance recipient costs ~130,285, roughly double) and 231,561 for a SunSwap V3 swap. This reproduces #834's table within ~$0.02 (USDT exact; SunSwap ±$0.01–0.02 from 3-decimal TRX-price rounding). (Minor ≤$0.02 gaps vs #834's published cells — e.g. 2025-Q3-pre USDT $4.39 vs $4.38, 2026-Q1 USDT $1.87 vs $1.86 and SunSwap $6.73 vs $6.71 — are TRX-price averaging precision over the same windows, not a method difference.)

Section 1 — Energy demand has plateaued

Total on-chain energy rose through 2025 (from a ~130B/day early-January trough to ~209B/day by Dec 2025) but has been flat across 2026 H1, oscillating in a ~190–203B/day monthly-average band with no directional trend (see the §2 monthly table; note weekly readings dipped to ~165B in a soft Feb). The post-reduction expansion has settled into a plateau.

Chart: Daily and Weekly Total Energy Consumption

Section 2 — The structural shift deepens: staking dominance and inflation

Monthly structural metrics, pre-proposal through 2026-06 (official energy-split and supply series):

Month	Total Energy (B/d)	Burn share %	Staking %	Annualized net supply Δ %
2025-06 (pre)	184.3	15.3	84.7	−0.87
2025-07 (pre)	184.8	15.0	85.0	−1.00
2025-08 (proposal)	185.0	13.5	86.5	−0.70
2025-09	199.2	14.9	85.1	+0.09
2025-10	208.9	15.5	84.5	−0.03
2025-11	208.4	12.5	87.5	+0.19
2025-12	209.1	12.8	87.2	+0.16
2026-01 (#834)	201.8	11.7	88.3	+0.30
2026-02	190.0	12.5	87.5	+0.32
2026-03	198.4	12.1	87.9	+0.29
2026-04	202.2	11.7	88.3	+0.29
2026-05	195.6	11.2	88.8	+0.37
2026-06	200.3	10.0	90.0	+0.44

Chart: Daily and Weekly Proportion of Energy Consumption by Staking

Direct-burn share of energy fell from 15.3% (Jun 2025) and 11.7% (Jan 2026, #834) to ~10.0% in Jun 2026, and staking now supplies ~90% of energy (by Jun 2026). Net TRX supply turned inflationary after the cut: −0.87% (Jun 2025, deflation) → +0.30% (Jan 2026, #834) → +0.44% (Jun 2026) annualized. The path was non-monotonic — a Mar–Apr plateau near +0.29%, then a sharp May–Jun rise to the Jun peak. In supply terms: daily TRX burn (~2.8M) no longer offsets gross issuance (~3.9M/day), leaving a net +~1.1M TRX/day and widening.

Note (issuance): #834 used ~5.06M/day, the old block+vote reward rate (16 + 160 TRX/block). A mid-2025 governance change cut that to 8 + 128 TRX/block, so on-chain issuance has been ~3.9M/day since (136 × ~28,800 blocks/day ≈ 3.92M, which matches TronScan's Generated series exactly). Using the current rate is what makes the burn-vs-issuance balance tip.

Note (burn-share basis): the Jun figure is ~10.0% on the ratio-of-monthly-sums method that reproduces #834's series (Jan 2026 = 11.7%). A mean-of-daily-ratios method reads 9.9% for Jun; we keep the #834-consistent basis, so Jun is ~10.0% (right at, not yet below, 10%).

Section 3 — Network economic footprint: fiat burn cost

Daily burn cost stayed low versus an "old-fee counterfactual" (post-cut burn × 210/100). Per-transaction cost is ~52% below the pre-cut level, straight from the 210 → 100 sun unit-price change; the USD figure tracks TRX price.

Chart: TRON Network Total Energy Consumption by Burning and Burn Cost (USD)

Section 4 — User-layer activation (small-value adoption)

Continuing #834's two published bands (warehouse transaction data, pre-proposal window vs through 2026-06):

• $1–10 (micro-payment): +65%, up from [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's +56% — small-value activation strengthened further. This is a clean like-for-like read: reproducing [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's Aug–Dec window on our transaction-level basis gives +56% for this band, matching [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's published figure, so the move to +65% is real.
• $10–1000 (core layer): ~+8%, roughly flat. [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834 published +11% for this band, but on a transfer-level count; reproduced on the same transaction-level basis we use, [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's window comes out ~+8.5%. So like-for-like the core band held about flat — it did not clearly decelerate; the apparent "+11% → +8%" is mostly a counting-basis gap, not a real slowdown.

So the picture is: activation strengthened in the smallest band and roughly held in the core — the gains concentrated rather than broadened. (A finer breakdown, beyond #834's two bands: a sub-$1 "dust" tier reversed from +15% to −17%, but it's the noisiest tier — batching, dust-attack decay, or a real drop are all plausible — so don't over-read it.) New-user inflow continued (raw new-user share of post-cut actives ≈ 93%; raw retention ≈ 45%).

Note (lifecycle): those two lifecycle numbers are raw (all addresses). #834's headline (new 69.3% / retention 62.6%) is on a classified user-layer (bots / Fan-Out / ephemeral excluded); reproducing those exactly for Q2 needs the classification step, which isn't done yet, so treat the raw figures as directional only — note raw retention (45%) is not comparable to #834's classified 62.6%. The two band figures above are user-layer-independent.

Insight / path forward

Net, the fee cut's effect on activity is positive but narrowing. The activation gains are real but concentrated: the $1–10 micro-payment band strengthened further (+56% → +65%) and has held up 10 months on, while the $10–1000 core band stayed roughly flat (+8%). So activity is deepening in the smallest band, not broadening across the fee base.

Meanwhile the inflation side #834 flagged has deepened (it was already +0.30% in Jan; it is +0.44% now). And #834's "volume-for-price" hope — that more transactions would make up for the lower per-unit price and keep burn up — has not held: even as USDT transactions grew (~+7% by count over the same before/after windows), total daily TRX burn fell from ~6.6M in mid-2025 to ~2.8M now, because staking has taken over the energy supply (burn share of energy 15% → 10%) and a shrinking slice of activity actually pays via burning. The volume didn't offset the loss — the burn path itself shrank.

So the balance has tilted further toward the structural concern, even though the headline activity story is still net-positive. The lever is unchanged: broadening the diversity of on-chain activity (DeFi, dApps, execution-heavy use) is what widens the direct-burn path. The practical read for governance is to gate any further fee reduction on a burn-path recovery first — i.e. not to cut again until diversified activity can absorb it — rather than to cut and hope volume compensates, which this quarter's data says it will not.

Data & method

• Sources: §1–§3 and the cost table are from official TronScan chart series (energy split, TRX supply/burn/issuance, prices), through 2026-07-13; §4 is from the on-chain transaction warehouse (transaction-level), through 2026-07-13.
• Validation before extending: the §2 structural table matches [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834 across the overlapping months within rounding (8 of 9 exact; Dec 2025 reads +0.16 vs [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's +0.15, a 0.155 rounding boundary), and the cost table matches the overlapping quarters within ~$0.02; §1/§2 charts match [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's originals. §4 continues [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's two published bands on a transaction-level basis (the warehouse runs ~9% below a transfer-level count on the smallest, most-batched transfers, so the $1–10 magnitude is directional though its direction is robust); §4 lifecycle matches the original study within ~3.4% on the 2025 window.
• Burn-share basis: ratio-of-monthly-sums (reproduces [Discussion] TRON Quarterly Transaction Fees Review — Q1 2026 #834's Jan 2026 = 11.7%); a mean-of-daily basis reads ~0.1–0.2pp lower.

[waynercheung]

Thanks for continuing this quarterly review. I independently checked the June figures against TRONSCAN, and the published totals are consistent.

One important clarification: energy_usage_total includes dynamic penalty Energy (energy_penalty_total), so it is useful for resource and burn accounting but is not a clean proxy for actual execution load. Could a future update split base Energy (total - penalty) from energy_penalty_total, ideally by contract or method?

Also, June's ~90% resource-covered share implies roughly 180.25B Energy/day covered by user and contract-deployer staked resources—very close to the network's nominal 180B staked-Energy budget (getTotalEnergyCurrentLimit, with adaptive Energy disabled). Given rolling resource recovery and per-account resource fragmentation, I would interpret this as high aggregate utilization rather than a strict daily ceiling.

This suggests the ~10% burn share may largely represent the residual resource shortfall after available staked resources are consumed, rather than an explicit user payment choice.

I agree that another fee reduction should wait until this decomposition is available.

[yanghang8612]

Thanks for the Q2 update. The recommendation to gate any further fee reduction on a “burn-path recovery” makes sense, but could this be translated into measurable review criteria?

For example, should recovery mean:

• daily burn / issuance returning to ≥1;
• net supply growth staying at or below zero for several consecutive months;
• burn share recovering above a specified threshold; or
• diversified DeFi/dApp activity reaching a defined level?

It would also be helpful to decompose the current burn decline into the effects of the lower unit price, total energy demand, and the staking-vs-burning payment mix. This would clarify whether the failure of the “volume-for-price” offset is mainly caused by the activity plateau or by activity increasingly moving away from direct burning.

Defining these thresholds would make future quarterly reviews more actionable and reduce ambiguity around when another fee adjustment could be considered.

[abn2357]

Thanks for continuing this quarterly review. I independently checked the June figures against TRONSCAN, and the published totals are consistent.

One important clarification: energy_usage_total includes dynamic penalty Energy (energy_penalty_total), so it is useful for resource and burn accounting but is not a clean proxy for actual execution load. Could a future update split base Energy (total - penalty) from energy_penalty_total, ideally by contract or method?

Also, June's ~90% resource-covered share implies roughly 180.25B Energy/day covered by user and contract-deployer staked resources—very close to the network's nominal 180B staked-Energy budget (getTotalEnergyCurrentLimit, with adaptive Energy disabled). Given rolling resource recovery and per-account resource fragmentation, I would interpret this as high aggregate utilization rather than a strict daily ceiling.

This suggests the ~10% burn share may largely represent the residual resource shortfall after available staked resources are consumed, rather than an explicit user payment choice.

I agree that another fee reduction should wait until this decomposition is available.

Regarding the first question: Penalty Energy is also part of the actual consumption. If it is split out, it would be detached from the actual business context. What would be the practical physical meaning of such a split?

Regarding the second question: Since energy does not actually recover linearly, theoretically within 24 hours it is possible to use 1.5 times the previous usage (with rigorous mathematical proof). This is also why 180.25B > 180B can occur. If utilization is high, it would not be a problem for staked energy usage to exceed 200B. Therefore, interpreting this as meaning that, within the current statistical period, users' actual staked energy consumption is approaching the ceiling—that seems to be fine.

[waynercheung]

@longgucun-yk
Agreed that penalty Energy is real, protocol-accounted consumption and must remain included in fee, burn, and resource accounting. The split is not intended to exclude it, but to answer a different question. Base Energy (total - penalty) represents the protocol-defined TVM cost of the executed path before the dynamic popularity multiplier, while penalty Energy is the additional dynamic surcharge rather than additional opcodes. The same execution path can therefore report different total Energy across maintenance cycles as energy_factor changes. Reporting both separates underlying metered execution from the dynamic surcharge, while total Energy remains the correct aggregate for resource consumption and its burn-paid portion remains the relevant input for tokenomics.

On recovery, agreed - the 180B value is not a hard UTC-day ceiling, and "high aggregate staked-resource utilization" is the better framing. Daily staked-Energy consumption divided by getTotalEnergyCurrentLimit may still be useful as a pressure ratio, but given resource recovery it should not be interpreted as a strict utilization percentage.

Also, because the billing path consumes the caller's available Energy before burning TRX, checking whether burning accounts retain unused allowance would be largely determined by construction. Resource fragmentation would instead require measuring unused capacity held elsewhere in the network. Is there an existing node-level metric for that, or would a breakdown of fully covered, mixed, and fully burned transactions be the most practical proxy?