高盛：2024代际增长报告：人工智能、数据中心和即将到来的美国电力需求激增（英文版）

VIP专享

3.0 村上 2024-07-05 30 3.35MB 35 页免费

侵权投诉

EQUITY RESEARCH | April 28, 2024 | 10:34PM EDT

Goldman Sachs does and seeks to do business with companies covered in its research reports. As a

result, investors should be aware that the ﬁrm may have a conﬂict of interest that could affect the

objectivity of this report. Investors should consider this report as only a single factor in making their

investment decision. For Reg AC certiﬁcation and other important disclosures, see the Disclosure

Appendix, or go to www.gs.com/research/hedge.html. Analysts employed by non-US afﬁliates are not

registered/qualiﬁed as research analysts with FINRA in the U.S.

The Goldman Sachs Group, Inc.

Carly Davenport

+1(212)357-1914

carly.davenport@gs.com

Goldman Sachs & Co. LLC

AI, data centers and the coming US power demand surge

Driven by AI, broader demand and a deceleration in the pace ofenergy eciency gains, global data center power demand is poised

to more than double by 2030 after being flattish in 2015-20. This growth is the primary catalyst alongside increasing

manufacturing/industrial production and broader electrication trends, to an acceleration in US electricity demand CAGR to 2.4%

through the end of the decade from 0% in the last decade. We believe supporting data center driven load growth will require

investmentby Utilities of $50bnin new power generation capacity.We assume a 60/40 split between gas and renewables, which

we expect to drive ~3.3 bcf/d incremental natural gas demand by 2030.While investor interest in the AI revolution theme is not

new, we believe downstream investment opportunities in utilities, renewable generation andindustrialswhose investment and

products will be needed to support this growth are underappreciated.

Brian Singer, CFA

+1(212)902-8259

brian.singer@gs.com

Goldman Sachs & Co. LLC

Neil Mehta

+1(212)357-4042

neil.mehta@gs.com

Goldman Sachs & Co. LLC

Brian Lee, CFA

+1(917)343-3110

brian.k.lee@gs.com

Goldman Sachs & Co. LLC

John Mackay

+1(212)357-5379

john.mackay@gs.com

Goldman Sachs & Co. LLC

See full list of authors inside

Note: The following is a redacted version of the original report published April 28, 2024 [43 pages].

Authors

Carly Davenport

+1 212 357-1914

carly.davenport@gs.com

Goldman Sachs & Co. LLC

Brian Lee, CFA

+1 917 343-3110

brian.k.lee@gs.com

Goldman Sachs & Co. LLC

Brendan Corbett

+1 415 249-7440

brendan.corbett@gs.com

Goldman Sachs & Co. LLC

Joe Ritchie

+1 212 357-8914

joseph.ritchie@gs.com

Goldman Sachs & Co. LLC

Jaskaran Jaiya

+1 332 245-7709

jaskaran.jaiya@gs.com

Goldman Sachs India SPL

Varsha Venugopal

+1 415 393-7554

varsha.venugopal@gs.com

Goldman Sachs & Co. LLC

Nick Cash

+1 212 357-6372

nick.cash@gs.com

Goldman Sachs & Co. LLC

Olivia Halferty

+1 801 212-7314

olivia.halferty@gs.com

Goldman Sachs & Co. LLC

Brian Singer, CFA

+1 212 902-8259

brian.singer@gs.com

Goldman Sachs & Co. LLC

John Mackay

+1 212 357-5379

john.mackay@gs.com

Goldman Sachs & Co. LLC

John Miller

+1 646 446-0292

john.y.miller@gs.com

Goldman Sachs & Co. LLC

Mark Delaney, CFA

+1 212 357-0535

mark.delaney@gs.com

Goldman Sachs & Co. LLC

Neil Mehta

+1 212 357-4042

neil.mehta@gs.com

Goldman Sachs & Co. LLC

Ati Modak

+1 212 902-9365

ati.modak@gs.com

Goldman Sachs & Co. LLC

Toshiya Hari

+1 646 446-1759

toshiya.hari@gs.com

Goldman Sachs & Co. LLC

Jerry Revich, CFA

+1 212 902-4116

jerry.revich@gs.com

Goldman Sachs & Co. LLC

Our views on key questions on the coming data center/AI power surge

How signiﬁcant will the power demand growth from AI/data centers be?

We forecast a 15% CAGR in data center power demand from 2023-2030, driving data

centers to make up 8% of total US power demand by 2030 from about 3% currently.

We now see a 2.4% CAGR in US power demand growth through 2030 from 2022

levels vs. ~0% over the last decade. Of the 2.4%, about 90 bps of that is tied to data

centers.

How much generation and overall capital investment will be required to

support data center power demand growth?

We estimate about 47 GW of incremental power generation capacity will be required

to support US data center power demand growth cumulatively through 2030, met with

about 60% gas and 40% renewable sources. We expect this to drive about $50 bn of

capital investment in US power generation capacity cumulatively through 2030.

Will there be bottlenecks in generation or transmission/interconnection

that constrain growth?

Initial conversations with utilities under coverage do not suggest material concerns on

supply chain on the generation kit or transmission side over the current ﬁve-year

planning period, though there have been concerns from investors that issues around

solar in recent years could reemerge. Further, lengthy interconnection queues remain

a challenge to connecting new projects to the grid, and expediting the

permitting/approval process for transmission projects will be key to alleviate it.

Elsewhere, we see similar potential growth constraints from natural gas transmission

infrastructure construction, speciﬁcally long-dated timelines, permitting challenges, and

environmental / landowner litigation. In our view, the most top of mind constraint for

natural gas is construction and permitting timelines where we see an average lag of

~4 years from the project announcement date to in-service date which means the

earliest capacity additions, if announced today, would not be in-service until ~2028.

If not, what will be the constraint to growth: AI budgets, demand or

neither?

The answer to this will be critical to the trajectory for AI power demand. The lack of any

constraint would be the most bullish for power consumption — i.e., if corporate

technology spending budgets are unlimited, then server shipments would remain

otherwise unchanged even as new generations of chips/servers come available that

cost more money but provide greater compute speed/consume more power (with

efﬁciencies). Demand as the constraint would be the most bearish, as, all else equal,

fewer more efﬁcient servers would be able to meet a ﬁxed level of demand. Our

estimates imply budgets are a constraint, i.e., buyers shift towards more productive

products within their budgets.

28 April 2024 3

Goldman Sachs Generational Growth

研报加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载研报到电脑，查找使用更方便

立即下载

标签： #美国 #人工智能

摘要：

EQUITY RESEARCH|April28,2024|10:34PMEDTGoldmanSachsdoesandseekstodobusinesswithcompaniescoveredinitsresearchreports.Asaresult,investorsshouldbeawarethattheﬁrmmayhaveaconﬂictofinterestthatcouldaffecttheobjectivityofthisreport.Investorsshouldconsiderthisreportasonlyasinglefactorinmakingtheirinvestment...

展开>> 收起<<

高盛：2024代际增长报告：人工智能、数据中心和即将到来的美国电力需求激增（英文版）.pdf

共35页,预览10页

还剩页未读，继续阅读