How large is the rotation into energy?
Large enough that it is the defining feature of the sponsor market this year. Energy accounted for roughly 25% of US private equity deal value through the first quarter of 2026, up from about 6% in 2025, and US energy private equity deal value is up 80.5% year to date against a total US private equity market down 37.5% quarter on quarter (Valuation Research Corporation, Q2 2026 Equity Markets Report on PitchBook data, published July 2026, through 31 March 2026; PitchBook Q2 2026 US PE Breakdown via Blue River Financial Group, published July 2026).
A second provider using a different method reaches the same place. Private equity investment in energy and natural resources stood at $90.2 billion at the end of the second quarter of 2026, on pace for a more than decade high, and the three largest US sponsor transactions of the quarter were all energy or digital infrastructure (KPMG, Pulse of Private Equity Q2 2026, on PitchBook data as of 30 June 2026).
The demand curve underneath it is the most specific forward number available anywhere. US data centre power demand is projected to rise from 31 gigawatts in 2025 to 41 gigawatts in 2026 and 66 gigawatts in 2027, on annual capacity additions of 36.3 gigawatts in 2027 against 8.5 gigawatts realized in 2025 (Goldman Sachs Research, 20 May 2026). Everything in this article follows from that step-up and from the grid work required to serve it.
The long-horizon spending numbers are forecasts rather than transaction data and should be read as such. Cumulative power infrastructure spending is projected at $25.0 trillion by 2050 with annual spending rising 76% to $1.1 trillion (PwC, energy, utilities and resources mid-year outlook, 23 June 2026). That is the thesis buyers are underwriting. It is not evidence of current activity.
Where does the middle-market opportunity actually sit?
In the services layer, because the capital is being deployed into assets and the assets have to be sited, permitted, connected, built, commissioned and maintained by somebody. Very little of that work is done by the owner of the asset.
The named categories are unglamorous and specific: site selection, grid interconnection feasibility, air permitting, generator emissions testing, water use modelling, electrical and mechanical contracting, commissioning, protective relay work, substation maintenance and the environmental consulting that surrounds all of it. Environmental consulting demand in particular is being driven by an infrastructure programme sized at roughly $3 trillion supported by around 100 gigawatts of new capacity additions (JLL via Capstone Partners, Industrial and Environmental Services M&A Update, 20 July 2026).
The transaction evidence in those services categories is running ahead of the market. Environmental services printed thirty-eight US transactions in June 2026, up 23% year over year against a roughly 17% decline in overall US deal count, with listed peers at a median 16.0x trailing EV/EBITDA as of 2 July 2026 (PitchBook data via Blue River Brief June 2026). Environmental, health and safety services reversed a two-year downtrend with sixty-one transactions year to date, up 15.1%, with sponsor platform formation up seven deals to nine (Capstone Partners, EHS Market Update, 30 June 2026).
One sub-sector is diverging and it is worth watching. Industrial and environmental services overall rose only 2.4% year over year to 167 transactions, with strategics taking 59.9% while sponsor add-ons fell 32.7% and platforms fell 20%, described as many willing and able buyers waiting to deploy capital (Capstone Partners, Industrial and Environmental Services M&A Update, 20 July 2026). Sidelined sponsor capital in a sector with rising demand usually resolves into a burst of platform formation once financing conditions clarify.
“The owners who benefit from this are not the ones building the plants. They are the ones who already do the permitting, the interconnection studies, the switchgear work and the commissioning, and who have spent a decade being told they are in a boring business. Their buyer list this year contains names that did not know they existed three years ago.”
Is the whole sector participating?
No, and the shape of the divergence has a name. Energy, utilities and resources deal volumes declined across all four sub-sectors in the first half of 2026 while value stayed resilient, a pattern PwC calls a K-shaped M&A market, with power and utilities leading on value largely because one proposed $67 billion merger accounted for more than half of sector deal value across the first five months (PwC, energy, utilities and resources mid-year outlook, 23 June 2026, on LSEG data to 31 May 2026).
Every energy headline this year is about the top of that K. Mid-market energy volumes are down and are forecast to stay soft even while value records fall, which is a materially different market from the one the headlines describe.
Upstream oil and gas shows the same divergence from the other direction. US upstream deal value fell to $9.1 billion in the second quarter of 2026, the third lowest quarterly total since 2020, down 76% quarter on quarter and 33% year over year, with more than 40% of that coming from a single lease sale (Enverus Intelligence Research, 5 August 2026). The same analysis frames the weakness as a bid-ask problem caused by crude volatility rather than a demand problem, and forecasts a materially busier second half.
The listed comparison shows how differently public and private capital are valuing the sector. Small-cap energy traded at 7.0x EV/EBITDA as of 30 June 2026 against 13.5x for the small-cap index overall, and energy was the only industry in that index to decline over the quarter (Valuation Research Corporation, Q2 2026 Equity Markets Report on S&P Capital IQ data). That gap is why public energy strategics are weak-currency buyers, which leaves the field to sponsors and to specialist capital.
| Segment | Current activity | Published pricing |
|---|---|---|
| US energy private equity | Roughly 25% of US private equity deal value through Q1 2026, from about 6% in 2025; deal value up 80.5% year to date; $90.2 billion of energy and natural resources investment at end-Q2 2026 | No private-market size-band multiple published for energy |
| Environmental services | 38 US transactions in June 2026, up 23% year over year against a roughly 17% decline in overall US deal count | Listed peers at a median 16.0x trailing EV/EBITDA as of 2 July 2026 |
| Environmental, health and safety services | 61 transactions year to date 2026, up 15.1%, with sponsor platforms up seven deals to nine | No sub-sector transaction multiple published |
| Industrial and environmental services | 167 transactions, up 2.4% year over year; strategics 59.9%, sponsor add-ons down 32.7%, platforms down 20% | No sub-sector transaction multiple published |
| Listed small-cap energy, for contrast | The only industry in the small-cap index to decline over the quarter, at minus 9.6% | 7.0x EV/EBITDA as of 30 June 2026 against 13.5x for the index overall |
What does a buyer underwrite in a transition services business?
Contract structure and policy exposure first, then the same backlog and workforce questions any project business faces. The distinguishing feature of this sector is that a meaningful part of the demand rests on incentives and rules that can change.
Policy dependence should be quantified rather than argued. What share of revenue depends on a specific incentive, credit, mandate or programme; what is the statutory or regulatory basis for it; when does it expire or come up for review; and what happens to the work if it does not survive. A business that can answer those four questions in writing is presenting a risk. A business that cannot is presenting an unknown, which prices worse.
Contract structure determines who carries the schedule risk, and schedule risk in this sector is real. Historically only about 72% of scheduled data centre capacity goes live on time, and the current adjustment is to roughly 60% over one year and about 50% over two (Goldman Sachs Research, 20 May 2026). A services business paid on milestones tied to a slipping programme has a working capital problem before it has a revenue problem.
On multiples, we will not give a private-market number for this sector, because none is published. Neither GF Data nor the other middle-market series produce size-band multiples for energy or infrastructure, and the honest available proxies are the listed small-cap energy figure of 7.0x, the listed environmental services median of 16.0x and the transaction averages published for adjacent industrial sub-sectors. None of those is a private energy transaction multiple, and anyone presenting one is estimating.
How could this thesis fail?
Two named ways, both currently visible. The first is that the financing gets ahead of the demand. The US high-yield market absorbed $31.9 billion of new bonds backing computing and data centre investment through 8 July 2026, all but $4.0 billion of it for new data centres, roughly 2.3 times the whole of 2025, with investment grade issuance at $218.0 billion against $80.5 billion in all of 2025 (Morningstar, 13 July 2026, originally published on PitchBook).
The same analysis reports signs of buy-side indigestion, with several issuers' paper trading wide of or below issue (same source). If that channel closes, the capital expenditure programme funding this demand slows before the demand itself changes, and the services businesses feel it first because they are paid last.
The second is schedule slippage compounding. The on-time adjustment above is already a fifty percent haircut over two years, and a services business whose forward plan assumes announced schedules rather than signed contracts is underwriting somebody else's optimism. That is a timing risk rather than a demand risk, but timing risk is what kills contractors.
There is a structural caution as well. Infrastructure fundraising hit a record low of $40.8 billion in the first half of 2026, below even the worst prior comparable, while individual funds set records, with one flagship closing at $19.2 billion (Infrastructure Investor, 14 July and 3 August 2026). A small number of very large, very motivated buyers is an excellent market for an asset owner and a difficult one for a mid-market manager. For a services business being sold, it means the buyer list is shorter and more concentrated than the sector's momentum suggests.
As of August 2026
Sources: Valuation Research Corporation, Q2 2026 Equity Markets Report, published July 2026, on PitchBook data through 31 March 2026, for energy at roughly 25% of US private equity deal value from about 6% in 2025, and on S&P Capital IQ data for small-cap energy at 7.0x EV/EBITDA as of 30 June 2026 against 13.5x for the small-cap index overall and energy as the only industry in that index to decline over the quarter at minus 9.6%; PitchBook Q2 2026 US PE Breakdown via Blue River Financial Group, published July 2026, for US energy private equity deal value up 80.5% year to date against a total US private equity market down 37.5% quarter on quarter, and for 38 US environmental services transactions in June 2026 at plus 23% year over year against a roughly 17% decline in overall US deal count with listed peers at a median 16.0x trailing EV/EBITDA as of 2 July 2026; KPMG, Pulse of Private Equity Q2 2026, on PitchBook data as of 30 June 2026, for $90.2 billion of private equity investment in energy and natural resources at end-Q2 2026, on pace for a more than decade high, and for the three largest US sponsor transactions of the quarter all being energy or digital infrastructure; Goldman Sachs Research, 20 May 2026, for US data centre power demand of 31, 41 and 66 gigawatts across 2025 to 2027, annual capacity additions of 36.3 gigawatts in 2027 against 8.5 gigawatts realized in 2025, and the on-time adjustment from a historical 72% to roughly 60% over one year and about 50% over two; PwC, Global M&A trends in energy, utilities and resources 2026 mid-year outlook, 23 June 2026, on LSEG data to 31 May 2026, for volumes declining across all four sub-sectors while value held, for the K-shaped characterization, for one proposed $67 billion merger accounting for more than half of sector deal value across the first five months, and for cumulative power infrastructure spending projected at $25.0 trillion by 2050 with annual spending rising 76% to $1.1 trillion; JLL via Capstone Partners, Industrial and Environmental Services M&A Update, 20 July 2026, for an infrastructure programme sized at roughly $3 trillion supported by around 100 gigawatts of new capacity additions, and the same Capstone update for 167 industrial and environmental services transactions at plus 2.4% with strategics at 59.9%, sponsor add-ons down 32.7% and platforms down 20%; Capstone Partners, EHS Market Update, 30 June 2026, for sixty-one EHS transactions year to date at plus 15.1% with sponsor platforms up seven deals to nine; Enverus Intelligence Research, 5 August 2026, for US upstream deal value of $9.1 billion in Q2 2026 at minus 76% quarter on quarter and minus 33% year over year, with more than 40% from a single lease sale, and for the forecast of a materially busier second half; Morningstar, 13 July 2026, originally published on PitchBook, for $31.9 billion of high-yield issuance backing computing and data centre investment through 8 July 2026 with all but $4.0 billion for new data centres at roughly 2.3 times all of 2025, for investment grade issuance of $218.0 billion against $80.5 billion in all of 2025, and for the reported signs of buy-side indigestion; Infrastructure Investor, 14 July 2026 and 3 August 2026, for first-half 2026 infrastructure fundraising at a record-low $40.8 billion and for a flagship fund closing at $19.2 billion. No middle-market publisher issues size-band transaction multiples for energy or infrastructure, so none is quoted here. Companion articles on this site cover specialty trades and backlog businesses, and the sectors where demand is structural rather than cyclical.