Cell tower and data center ABS (digital infrastructure)

Does your product fit here?

Digital infrastructure ABS covers two distinct but related asset types: cell towers and data centers. Both generate long-term contracted cash flows from essential infrastructure, but the underwriting, structure, and investor base differ materially.

Cell towers

Tower securitization captures lease revenue from wireless carriers (AT&T, Verizon, T-Mobile) who rent antenna space on tower infrastructure. The core product is macro towers, typically 100-400 feet tall, hosting multiple carrier tenants.

What qualifies:

Multi-tenant tower portfolios with master lease agreements (MLAs)
Single-tenant towers with colocation rights
Ground lease ownership or long-term ground lease control
Rooftop installations with long-term site access

What doesn’t fit here:

Fiber network securitization: different revenue model (capacity-based, not lease-based), different structure
Wireless spectrum: intangible asset requiring specialized securitization treatment
Telecom equipment: this is equipment ABS, not infrastructure ABS

Edge cases:

Small cells and distributed antenna systems (DAS): emerging category, treated as mini-towers but with different economics (lower revenue per site, higher density requirements)
In-building wireless: harder to finance standalone; typically bundled with building real estate or as part of larger tower portfolio

Data centers

Data center securitization captures lease revenue from enterprise tenants and cloud providers who rent space, power, and connectivity.

Three distinct models, each financed differently:

Model	Tenant Profile	Typical Lease	Financing Approach
Hyperscale	Single tenant (AWS, Azure, Google)	10-20 years, triple-net	Sale-leaseback, WBS, private placement
Wholesale colocation	5-20 large tenants	3-10 years	WBS, term loan, CMBS
Retail colocation	50-500+ tenants	1-3 years	WBS, warehouse, term loan

What qualifies:

Stabilized facilities with 85%+ occupancy
Diversified tenant base (retail/wholesale colo) or investment-grade anchor tenant (hyperscale)
Adequate power and cooling capacity for contracted load
Carrier-neutral interconnection

What doesn’t fit here:

Development-stage data centers: not yet stabilized, requires equity or construction financing
Edge computing facilities: too small and too new for institutional securitization
Bitcoin mining facilities: highly volatile tenant credit, specialized power requirements, limited financing market

How lenders will classify you

Your capital source options depend on scale, tenant quality, and contract duration:

Profile	Capital Options
Large tower portfolio (1,000+ towers), investment-grade tenants	Public ABS, 144A, insurance private placement
Mid-size tower portfolio (100-1,000 towers)	Private placement, term loan, warehouse
Single hyperscale data center, 15-year lease	Sale-leaseback, insurance private placement
Multi-tenant colo, diversified tenants	WBS, CMBS, warehouse facility
Smaller or development-stage	Equity, mezzanine, construction loan

Market benchmarks and comps

Cell tower performance metrics

These are the metrics capital providers use to evaluate tower portfolios. If you can’t provide these, you’re not ready for institutional financing.

Metric	Strong Portfolio	Acceptable	Raises Flags
Weighted avg. remaining lease term	> 7 years	5-7 years	< 5 years
Colocation ratio (tenants/tower)	> 2.0x	1.5-2.0x	< 1.5x
Annual tenant churn	< 2%	2-3%	> 3%
Investment-grade tenant concentration	> 90%	75-90%	< 75%
Ground lease ownership	> 50% owned	30-50% owned	< 30% owned
Annual lease escalator	3%+	2-3%	< 2%

What “good” looks like: Crown Castle and American Tower portfolios show tenant churn below 1.5%, colocation ratios of 2.0-2.5x, and virtually zero tenant defaults because all major carriers are investment-grade.

What raises flags: Single-tenant towers with no colocation potential, towers dependent on a single carrier facing credit stress, ground leases expiring within 5 years with no renewal locked.

Data center performance metrics

Metric	Hyperscale	Wholesale Colo	Retail Colo
Typical lease term	10-20 years	5-10 years	1-3 years
Renewal rate	95%+	85-95%	75-85%
Tenant churn	< 1%	3-5%	5-10%
PUE (power efficiency)	1.2-1.4	1.3-1.6	1.4-1.8
Contracted power load	100% utilized	70-90%	60-80%

PUE (Power Usage Effectiveness) measures how efficiently a data center uses power. A PUE of 1.5 means the facility uses 1.5 watts of total power for every 1 watt delivered to computing equipment. The best hyperscale facilities run below 1.2; older retail colo facilities can exceed 2.0.

Pricing benchmarks (mid-2026 environment)

Cell tower ABS (rated):

AAA tranche (5-7 year WAL): SOFR + 80-120 bps
A-rated tranche: SOFR + 140-180 bps
BBB tranche: SOFR + 200-275 bps

Data center financings:

Hyperscale sale-leaseback (investment-grade tenant): 5.5-6.5% fixed rate
Wholesale colo WBS AAA: SOFR + 100-150 bps
Private placement (IG-rated): SOFR + 150-225 bps
Warehouse facility: SOFR + 200-300 bps

Leverage metrics:

Tower WBS: 5-7x EBITDA typical; investment-grade-rated deals can push to 7-8x
Data center WBS: 4-6x EBITDA; hyperscale sale-leaseback can exceed 8x with long lease term

What lenders and investors focus on

Cell towers: the five factors that drive credit decisions

1. Carrier Tenant Mix and Credit Quality

Your tenant roster determines 80% of the credit analysis. All three major U.S. carriers (AT&T, Verizon, T-Mobile) are investment-grade. If your portfolio has significant non-investment-grade tenant exposure (regional carriers, MVNOs), expect:

Lower advance rates (5-10% haircut)
Higher spreads (25-50 bps)
Concentration limits on sub-IG tenants

Note: Most institutional investors require 75%+ of revenue from investment-grade tenants. Below that threshold, you’re in a different market.

2. Colocation Economics

A tower with one tenant generates revenue but has no cushion. A tower with 2-3 tenants generates higher revenue and provides diversification. Incremental tenants after the first add revenue at 85%+ margin because the tower infrastructure is already built.

Colocation Ratio	Implication
1.0x (single tenant)	Maximum tenant concentration risk; pricing haircut
1.5x	Acceptable diversification
2.0x+	Strong economics; preferred by investors
3.0x+	Premium portfolio; maximum advance rate

3. Ground Lease Control

You either own the land beneath the tower or you rent it. This matters enormously:

Owned ground: Best case. You control the site indefinitely.
Long-term ground lease (20+ years remaining): Acceptable. Factor ground rent into cash flow.
Short-term ground lease (< 10 years): Problem. Renewal risk creates uncertainty; may require ground lease buyout reserves.
Expiring ground lease (< 5 years): Deal-breaker without locked-in renewal or buyout agreement.

Ground rent typically runs 15-25% of tower revenue. If your ground leases are below market, there’s upside; if above market, you’re overpaying and margins compress.

4. Technology Transition Risk

5G deployment is driving carrier capital expenditure, but it’s also changing network architecture. Small cells and distributed antenna systems handle density; macro towers handle coverage. Towers in urban areas face more small-cell competition; rural macro towers remain essential.

Questions investors will ask:

What percentage of your portfolio is in high-density urban areas vs. suburban/rural?
Have any carriers signaled lease non-renewal due to network architecture changes?
Are your towers 5G-ready (structural capacity for additional equipment)?

5. Geographic Distribution

Single-market concentration creates correlated risk. A portfolio concentrated in one metro area faces:

Carrier-specific market decisions (network rationalization after M&A)
Natural disaster exposure
Regulatory/zoning risk

Institutional investors prefer national diversification. Single-state concentration above 25% will trigger questions; above 40% will affect pricing.

Data centers: the five factors that drive credit decisions

1. Tenant Concentration and Credit Quality

The tradeoff is binary: hyperscale facilities have concentrated but extremely creditworthy tenants; colocation facilities have diversification but weaker individual tenant credit.

Tenant Type	Credit Quality	Concentration Risk	Typical Approach
Hyperscale (AWS, Azure, Google)	AAA/AA equivalent	Single-tenant	Long lease term mitigates
Enterprise (Fortune 500)	BBB to A	10-20% per tenant	Concentration limits
SMB (retail colo)	Unrated	< 5% per tenant	Diversification credit

For hyperscale deals, the credit analysis is essentially a tenant credit analysis plus real estate. For colo deals, you need diversification: no single tenant above 15-20% of revenue.

2. Power Availability and Cost

Data centers are power-constrained. The critical questions:

Contracted power capacity: How much utility power is available to the facility?
Utilized vs. available: If you’re at 95% utilization, there’s no room for growth; if at 60%, there may be lease-up risk.
Power cost: What’s your blended electricity rate? Below $0.05/kWh is competitive; above $0.10/kWh is challenging.
Renewable energy: Increasingly required by enterprise tenants; PPAs and RECs matter for tenant retention.

Power density is increasing rapidly due to AI/ML workloads. Traditional data centers support 5-10 kW per rack; AI training requires 30-50+ kW. If your facility can’t support high-density computing, it may face obsolescence risk.

3. Interconnection and Network Density

Colocation value depends on connectivity. Carrier-neutral facilities with multiple network options command premium rents. Metrics that matter:

Number of network providers on-site
Direct cloud on-ramps (AWS Direct Connect, Azure ExpressRoute)
Peering exchange presence
Cross-connect revenue as % of total

A facility with 50+ networks and direct cloud connectivity is more defensible than one with limited options. Tenants won’t leave a well-connected facility easily.

4. Physical Infrastructure Quality

Capital providers will assess:

Cooling redundancy: N+1 or 2N cooling systems
Power redundancy: UPS, generator backup, utility feed diversity
PUE: Lower is better (1.2-1.4 is excellent; > 1.8 is dated)
Age and deferred maintenance: Facilities over 15 years old require capex reserves
Certifications: SOC 2, HIPAA, PCI-DSS compliance

5. Lease Expiration Schedule

Clustered lease expirations create refinancing risk. A portfolio with 40% of leases expiring in a single year has a problem. Investors want:

Staggered expirations (no more than 20% in any year)
Weighted average lease term > 5 years for wholesale, > 10 years for hyperscale
Strong renewal history

Typical structures used

Whole business securitization (WBS)

The dominant structure for large tower portfolios. The operating company transfers tower assets to an SPV; the SPV issues rated debt secured by all cash flows.

How it works:

Operating company contributes towers to a bankruptcy-remote SPV
SPV issues multiple tranches of rated debt (typically A through BB)
Operating company retains equity interest and continues managing towers
Cash flows service debt in priority order; excess returns to equity

Typical terms:

Advance rate: 55-70% of enterprise value (tower portfolios can push higher due to stable cash flows)
Leverage: 5-7x EBITDA
Amortization: 1-2% annual, with bullet at maturity
Tenor: 5-7 year tranches, 10-15 year legal final
Covenants: DSCR (typically 1.75x minimum), leverage ratio, tenant concentration limits

Who uses this: Crown Castle, American Tower, SBA Communications have all issued tower WBS. Data center operators including CoreSite and QTS have used similar structures.

Sale-leaseback

Common for single-tenant hyperscale facilities. The operator sells the real estate to an investor and leases it back on a long-term triple-net basis.

How it works:

Data center operator sells facility to investor (real estate fund, REIT, infrastructure fund)
Operator signs 15-25 year NNN lease with annual escalators
Investor takes real estate residual risk; operator retains equipment and operations

Typical terms:

Cap rate: 5-7% (hyperscale with long lease); 6-8% (shorter lease or weaker tenant)
Lease term: 10-25 years
Escalators: 2-3% annual
Tenant credit: Investment-grade required for best pricing

Who uses this: Cloud providers (AWS, Google, Microsoft) and enterprise data center operators monetize facilities this way. Buyers include Digital Realty, Brookfield, KKR, and insurance company real estate portfolios.

Private placement / rated notes

For operators who want term funding with insurance capital but don’t want public issuance.

How it works:

Operator (or SPV) issues medium-term notes (10-15 years)
Notes rated by one or two agencies
Insurance companies purchase for NAIC-1 or NAIC-2 capital treatment

Typical terms:

Size: $100M-$500M
Tenor: 10-15 years, matching insurance ALM
Pricing: 5.5-7.5% fixed (spread to comparable Treasury)
Covenants: Leverage ratio, interest coverage, tenant concentration

Who uses this: Mid-size tower operators, data center platforms seeking long-term capital without public market execution.

Warehouse facility

For aggregation before term-out or for platforms too small for term securitization.

Typical terms:

Advance rate: 50-65%
Pricing: SOFR + 200-300 bps
Tenor: 2-3 year revolving
Minimum size: $50M committed

Asset-class-specific structural features

Cell tower structures

Cash trap triggers: Tower WBS typically includes DSCR-based triggers:

DSCR < 1.75x: Cash trapping begins (excess spread held in reserve)
DSCR < 1.50x: Rapid amortization triggered
DSCR < 1.25x: Potential acceleration event

Tenant concentration limits: Single-carrier concentration typically capped at 40-50% of revenue. If one carrier exceeds this, excess revenue may be excluded from borrowing base.

Ground lease reserves: Structures often require reserves for ground lease renewal or buyout if leases expire within the facility term.

Technology upgrade requirements: Covenants may require maintaining 5G-ready capacity or funding equipment upgrades.

Colocation incentives: Some structures include provisions encouraging colocation (reduced advance rate haircut as colocation increases).

Data center structures

Power purchase agreement (PPA) assignment: For facilities with third-party power contracts, the PPA may need to be assigned to the SPV or the lender may require step-in rights.

Capex reserves: Required reserves for cooling upgrades, power expansion, and deferred maintenance. Typically sized at $10-25/kW annually.

Tenant improvement allowances: Facilities that fund TI for new tenants need reserves or covenant carve-outs.

Interconnection revenue: May be treated separately (carved out or subordinated) from base lease revenue.

Environmental compliance: Diesel generator permits, water usage limits (for evaporative cooling), and carbon disclosure requirements increasingly included in covenants.

Rating agency treatment

S&P global

For tower ABS, S&P focuses on:

Tenant credit quality: Investment-grade tenants provide credit floor
Contract duration: Weighted average remaining lease term
Technology risk: Long-term view on macro tower relevance
Ground lease control: Owned vs. leased, remaining term

S&P typically sizes to 2.5-3.5x stressed cash flow decline for AAA, assuming tenant churn acceleration and lease non-renewal in stress.

Moody’s

Moody’s emphasizes:

Essentiality of infrastructure: Carriers cannot easily relocate; replacement cost is high
Operator dependency: Management quality, track record, financial strength
Geographic diversification: Concentrated portfolios receive haircuts

Moody’s generally requires higher subordination than S&P for tower deals, particularly where ground lease exposure is significant.

Fitch

Fitch provides detailed analysis on:

Ground lease renewal probability: Models lease-by-lease renewal likelihood
Carrier industry dynamics: M&A impact, network sharing arrangements
Cash flow stability: Historical volatility in revenue and EBITDA

KBRA and DBRS

Both have rated digital infrastructure deals. KBRA is more flexible on track record for emerging operators. DBRS has methodology for both tower and data center ABS.

Typical enhancement levels

Structure	AAA Subordination	Total Credit Enhancement
Tower WBS (diversified portfolio)	15-20%	20-30%
Tower WBS (concentrated)	20-30%	30-40%
Data center WBS (multi-tenant)	20-30%	30-40%
Data center (hyperscale, long lease)	10-15%	15-25%

Diligence focus areas

Tower portfolio diligence

Lease abstracts: Capital providers will require lease-level detail for every carrier tenant:

Lease commencement and expiration dates
Rental rate and escalation terms
Renewal options and notice periods
Termination rights and penalties
Amendment history

Ground lease analysis:

Remaining term and renewal options
Current rent vs. market rent
Landlord consent requirements for assignment
Purchase option terms (if any)

Site-level diligence:

Zoning compliance verification
Structural engineering reports (capacity for additional tenants)
Environmental phase I reports
Title insurance or ownership verification

Colocation analysis:

Structural capacity for additional tenants
Market assessment of colocation opportunity
Historical colocation success rate

Regulatory and permitting:

FAA registration status
FCC compliance
Local permitting and height restrictions

Data center diligence

Power and cooling:

Utility capacity documentation
PPA and power supply agreements
Cooling system specifications and redundancy
Generator and UPS testing records
PUE history and improvement plans

Interconnection:

Network provider agreements
Cross-connect inventory and pricing
Cloud on-ramp documentation
Peering arrangements

Tenant documentation:

Lease abstracts with financial terms
Tenant credit analysis
Renewal history and retention rates
TI obligations and funding status

Physical inspection:

Facility tour and condition assessment
Deferred maintenance identification
Security and access control systems
Compliance certifications (SOC 2, etc.)

Environmental:

Diesel generator permits and testing
Water usage rights (if evaporative cooling)
Hazardous materials handling
Carbon disclosure and sustainability reporting

Common diligence findings and what they mean

Finding	Implication
Ground leases expiring within 5 years	Requires renewal verification or reserve; may reduce advance rate 5-10%
Single-tenant towers > 30% of portfolio	Concentration limit trigger; may require exclusion from borrowing base
PUE > 1.8	Dated facility; requires capex reserve or advance rate haircut
Tenant lease expirations clustered	Refinancing risk; may require lease extension as condition
Missing or incomplete lease abstracts	Due diligence failure; cannot close until resolved

Active participants

Tower operators / originators

Public tower REITs (largest U.S. portfolios):

Crown Castle (CCI): 40,000+ U.S. towers, active WBS issuer
American Tower (AMT): 40,000+ U.S. towers (220,000+ globally), investment-grade corporate issuer
SBA Communications (SBAC): 17,000+ U.S. towers, active WBS issuer

Private tower operators:

Phoenix Tower International
Vertical Bridge
Landmark Dividend
Tillman Infrastructure

Data center operators

REITs and large operators:

Equinix: Largest global colo operator, investment-grade corporate
Digital Realty: Second-largest globally, both hyperscale and colo
QTS (owned by Blackstone): Large U.S. portfolio
CyrusOne (owned by KKR/GIP): Enterprise-focused
CoreSite: Dense urban locations
DataBank: Regional colo platform

Hyperscale operators (build/own):

AWS, Microsoft Azure, Google Cloud (primarily own rather than lease)
Meta, Apple, Oracle (significant owned capacity)

Capital providers

Banks (warehouse and term underwriting):

JPMorgan, Bank of America, Goldman Sachs, Deutsche Bank, Barclays
Morgan Stanley, Wells Fargo, Citi

Insurance capital:

MetLife, Prudential, TIAA, Athene, Lincoln Financial
Typically seeking AAA or A-rated tranches with 10+ year duration

Infrastructure funds:

Macquarie Infrastructure, Brookfield, KKR, EQT
Global Infrastructure Partners, IFM Investors
DigitalBridge (specialist in digital infrastructure)

Credit funds:

Apollo, Ares, Blackstone Credit, Blue Owl
Typically for higher-yielding tranches or private placements

ABS underwriters / bookrunners

For rated tower ABS: Barclays, Deutsche Bank, Goldman Sachs, JPMorgan, Morgan Stanley, RBC, Wells Fargo

Legal counsel

Issuer/operator side: Latham & Watkins, Simpson Thacher, Davis Polk, Kirkland & Ellis
Underwriter/lender side: Milbank, Cleary Gottlieb, Cravath, Sullivan & Cromwell
Trustee counsel: Chapman and Cutler, Seward & Kissel

Trustees

U.S. Bank, Wilmington Trust, Deutsche Bank Trust Company Americas

Red flags and off-market characteristics

Tower portfolio red flags

These will widen your pricing, reduce advance rates, or kill the deal:

Tenant-related:

Single-tenant towers exceeding 40% of portfolio with no colocation path
Non-investment-grade carriers exceeding 25% of revenue
Any carrier with public credit deterioration (downgrade trajectory)
Carrier churn exceeding 3% annually

Ground lease-related:

Ground leases expiring within 5 years without locked renewal or buyout
Ground rent exceeding 30% of tower revenue (margin compression)
Landlord disputes or litigation
Below-market purchase options held by third parties

Technology/market-related:

High concentration in dense urban areas facing small-cell displacement
Towers lacking 5G upgrade capacity (structural limitations)
Markets with carrier network sharing (reduces colocation opportunity)

Financial/structural:

EBITDA margin below 50% (operational inefficiency)
Historical churn exceeding market norms without explanation
Leverage exceeding 7x without investment-grade rating

Data center red flags

Tenant-related:

Single-tenant hyperscale exposure without 15+ year lease
Top 3 tenants exceeding 60% of revenue (retail/wholesale colo)
Significant sub-investment-grade tenant concentration
Lease expirations clustered (> 25% in any single year)
Declining renewal rates (below 80% for colo)

Infrastructure-related:

PUE exceeding 1.8 (dated cooling infrastructure)
Power capacity constraints (> 90% utilized with no expansion path)
Single utility feed without backup
Deferred maintenance exceeding $50/kW

Market-related:

Location in oversupplied market (Northern Virginia, certain secondary markets)
Limited interconnection options (carrier-neutral status at risk)
No direct cloud on-ramps for enterprise-focused facility

Financial/operational:

Occupancy below 80% for facilities older than 3 years
Negative net absorption (more departures than new leases)
Operating expense ratio exceeding industry norms
Management team turnover

Common themes: critical infrastructure

Both cell towers and data centers share characteristics that make them attractive for securitization:

Long-term contracted cash flows

Wireless carriers sign 5-10 year tower leases with multiple renewal options. Hyperscale tenants sign 10-20 year data center leases. Enterprise data center tenants sign 3-10 year agreements. Built-in escalators (2-3% annually) provide inflation protection.

Essential infrastructure with high switching costs

Carriers cannot easily relocate tower equipment. The cost of moving antennas, reconfiguring networks, and potentially degrading coverage is prohibitive. Similarly, data center tenants face significant migration costs: physical server relocation, network reconfiguration, interconnection re-establishment, and application downtime.

This “stickiness” translates to high renewal rates and predictable cash flows.

Technology obsolescence risk

Both asset classes face long-term questions about network architecture evolution:

Towers: 5G densification may shift value from macro towers to small cells in urban areas. However, macro towers remain essential for coverage (rural areas, highways) and capacity (backhaul for small cells). The risk is concentrated in dense urban portfolios.

Data centers: AI/ML workloads require dramatically higher power density than traditional enterprise computing. Older facilities may struggle to support 30-50+ kW per rack. The risk is concentrated in aging facilities without upgrade paths.

Capital intensity

Both require ongoing investment:

Towers: Ground lease payments, structural maintenance, 5G equipment support
Data centers: Cooling upgrades, power infrastructure, tenant improvements

Structures must account for required capex through reserves or covenant carve-outs.

ESG considerations

Increasingly scrutinized by institutional investors:

Energy consumption: Data centers consume 1-2% of global electricity; pressure for renewable sourcing
Carbon footprint: Both asset types face disclosure requirements and sustainability targets
Water usage: Evaporative cooling in data centers consumes significant water in some designs
Land use: Tower and data center development face community opposition in some locations

Operators with strong ESG profiles (renewable power, water-efficient cooling, carbon-neutral commitments) may access cheaper capital as sustainable finance mandates grow.