This article is a work in progress. If you have any questions, thoughts, or corrections, contact us.

Market Intelligence

Market data and public datasets

Market data and public datasets

Market intelligence separates informed capital allocation from guesswork. Whether you are an originator benchmarking your performance or a capital provider evaluating deals, knowing where to find data and how to use it efficiently is a core competency.

This guide maps the ABF data landscape: what is available for free, what requires a subscription, what each source is actually useful for, and how to build a practical intelligence workflow without overspending on data you will never use.

Free public data sources

You can build a surprisingly complete market picture using free sources. The tradeoff is time: free data requires more work to extract and clean.

SEC EDGAR

The SEC’s Electronic Data Gathering, Analysis, and Retrieval system is the definitive source for public ABS deal data.

What you can find:

  • ABS-EE filings: Loan-level data on every public ABS deal (consumer loans, auto, student loans, credit cards). This includes borrower characteristics, loan terms, and performance history.
  • 10-D reports: Monthly servicer reports with delinquency rates, prepayment speeds, loss amounts, and pool factor.
  • Prospectuses (424B2): Deal structure, waterfall mechanics, risk factors, and legal terms.
  • Shelf registrations (S-3): Issuer programs that signal upcoming deal activity.

Access: Free at sec.gov/cgi-bin/browse-edgar

Limitations:

  • Only public issuers. Private deals (most warehouse facilities, many smaller originators) are not filed.
  • ABS-EE files are XML format. You need to parse them programmatically or use a third party.
  • Filing lag of 15-45 days from deal close.

How to use it effectively:

  1. Use SEC’s full-text search with the issuer name or deal name to find specific filings.
  2. For ABS-EE, write a simple Python script (pandas + lxml) to parse XML into usable tables, or use services like Recursiv or dv01 that provide parsed data.
  3. Track 10-D reports monthly for benchmark deals in your asset class. Set up email alerts for new filings.
  4. Read prospectuses for deals similar to what you are structuring. The waterfall mechanics section tells you what subordination levels and triggers are market standard.

Note: SEC EDGAR is the most underutilized free resource in ABF. Most practitioners complain about data access while ignoring the loan-level data sitting in ABS-EE files.

FINRA TRACE

TRACE (Trade Reporting and Compliance Engine) captures secondary market trading in fixed income securities, including ABS.

What you can find:

  • Trade prices and volumes for publicly traded ABS
  • Historical price trends for specific CUSIPs
  • Dealer activity patterns

Access: Free at finra-markets.morningstar.com/BondCenter

Limitations:

  • Only public securitizations. No private deal pricing.
  • Some trades are reported with delays or masked (particularly large trades to protect dealer identity).
  • You see executed trades, not bid/offer spreads.

How to use it effectively:

  1. Look up specific CUSIPs to see where a bond has traded recently.
  2. Download historical data in CSV format for analysis.
  3. Use trading activity as a signal of market liquidity. Heavy volume may indicate stress or rebalancing.

FFIEC and federal reserve data

Bank regulatory filings provide insight into warehouse lending capacity and charge-off trends.

Key sources:

FFIEC Central Data Repository (cdr.ffiec.gov)

  • Bank call reports with balance sheet exposure by asset category
  • Schedule RC-S shows bank securitization and warehouse lending activity

Federal Reserve Statistical Releases:

  • H.8: Consumer credit statistics (total credit card, auto, student loan balances)
  • G.19: Consumer credit outstanding by category
  • Z.1 (Flow of Funds): Household and business debt composition
  • SLOOS (Senior Loan Officer Opinion Survey): Whether banks are tightening or loosening credit standards

FRED (fred.stlouisfed.org)

  • Economic time series with downloadable data and APIs
  • Charge-off rates, delinquency rates by loan type

Limitations:

  • Data is aggregated, not deal-level.
  • Quarterly lag on most reports.
  • Categories do not always map cleanly to ABF asset classes (e.g., “other consumer loans” lumps many things together).

How to use it effectively:

  1. Track charge-off rates in FRED for your asset class to benchmark your loss expectations against industry averages.
  2. Read SLOOS quarterly to understand whether warehouse capacity is expanding or contracting.
  3. Use Flow of Funds data to size total addressable markets for different asset classes.

GSE loan performance data

Freddie Mac and Fannie Mae publish loan-level performance data on millions of mortgages.

Access:

What you can find:

  • Loan characteristics at origination
  • Monthly performance (current, delinquent, prepaid, modified, defaulted)
  • Millions of observations for statistical modeling

Limitations: Only agency-eligible mortgages. Not directly applicable to most ABF asset classes, but useful if you are doing consumer credit modeling and need prepayment or default data to calibrate assumptions.

Paid sources provide cleaner data, real-time access, and analytical tools. The cost varies from $5,000/year for basic services to $100,000+ for comprehensive platforms.

Bloomberg terminal

Bloomberg is the default terminal for structured finance professionals. If you are a capital provider doing ABF at any scale, you will likely need it.

Relevant functions:

  • ABS page (ABS <GO>): Searchable database of public ABS deals with pool statistics and pricing
  • TRACE integration: Real-time and historical secondary trading
  • CMBS analytics: Property-level data, loan status, watchlist
  • CLO analytics: Portfolio composition, OC/IC test levels, manager performance
  • SRCH function: Filter deals by issuer, asset class, rating, vintage

Cost: $24,000-27,000/year per terminal

Good for: Daily trading, secondary market pricing, broad asset class coverage, access to research

Limitations:

  • Private deal coverage is limited
  • Depth varies by asset class (CMBS and CLO are comprehensive, esoteric ABF is sparse)
  • Steep learning curve for structured products functions

How to use it effectively:

  1. Learn SRCH ABS <GO> to filter deals by your criteria.
  2. Use DES <GO> on any CUSIP for deal details and pool statistics.
  3. Ask your Bloomberg rep for a structured finance training session. They offer free training and it is worth the time.
  4. Use the MSG function to contact the help desk. Bloomberg support is actually knowledgeable about structured products.

Intex

Intex is the industry standard for cash flow modeling in structured products. If you need to stress test a deal or understand waterfall mechanics in detail, this is the tool.

What you can do:

  • Load deal structures with full waterfall logic
  • Run cash flow projections under various prepayment, default, and recovery scenarios
  • Track historical performance against original projections
  • Export cash flows to Excel for custom analysis

Cost: $30,000-100,000+/year depending on coverage (CMBS, CLO, ABS, RMBS) and number of seats

Good for: Detailed deal analysis before investment, stress testing, relative value comparisons across deals

Limitations:

  • Primarily covers public deals. Private deal modeling requires manual setup.
  • Complex interface with a significant learning curve.
  • Cost is prohibitive for smaller firms.

How to use it effectively:

  1. Start with the deals you are actively evaluating. Do not try to learn every feature at once.
  2. Download CDI files (deal models) for your target deals. They are updated regularly.
  3. Build standard stress scenarios (base case, moderate stress, severe stress) and apply consistently across deals.

Trepp

Trepp specializes in CMBS and CLO analytics with loan-level granularity.

What you can find:

  • CMBS: Property-level data, loan performance, watchlist status, special servicing
  • CLO: Portfolio composition, credit metrics, manager benchmarking
  • Alerts: Notifications on specific deals or properties

Cost: $20,000-50,000+/year depending on coverage

Good for: CMBS and CLO specialists who need property-level or loan-level data

Limitations:

  • Minimal coverage of consumer ABS
  • Focused on commercial real estate and corporate credit

How to use it effectively:

  1. Subscribe to TreppWire (their daily newsletter). The free summary version provides market color.
  2. Use property-level data for CRE CLO due diligence. You can track individual properties in the portfolio.

LCD / pitchbook

LCD (now part of PitchBook) covers leveraged loan and private credit markets.

What you can find:

  • Leveraged loan pricing and trading levels
  • CLO arbitrage economics (loan spread vs. liability cost)
  • Middle market loan statistics
  • Private credit deal flow and terms

Cost: $15,000-40,000+/year

Good for: CLO managers, leveraged loan investors, private credit funds

Limitations: Focused on corporate credit, not consumer ABF

Dv01

dv01 is a consumer loan analytics platform with deep data on marketplace lending and consumer credit.

What you can find:

  • Loan-level performance data from major originators (with their permission)
  • Cohort analysis and vintage curves
  • Benchmark comparisons across originators

Cost: $50,000-150,000+/year

Good for: Consumer loan investors, marketplace lending specialists, originators benchmarking their performance

Limitations: Focused on consumer unsecured and student loans. Limited coverage of commercial ABF.

Finsight

Finsight tracks ABS new issue flow and secondary market color.

What you can find:

  • New issue pricing and deal terms
  • Secondary market trading commentary
  • Calendar of upcoming deals

Cost: Subscription tiers starting around $5,000/year

Good for: Tracking new issue flow, understanding where deals are pricing, market commentary

Recursiv

Recursiv parses ABS-EE data into usable formats, saving you the work of parsing XML yourself.

What you can find:

  • Deal-level and pool-level metrics from SEC filings
  • Custom screening and alerts
  • Time series performance data

Cost: Varies by coverage

Good for: Systematic ABS analysis, avoiding manual EDGAR work

Rating agency data

Rating agencies publish valuable data on default rates, rating transitions, and deal-level analysis. Access ranges from free (basic ratings) to expensive (full research subscriptions).

S&P global

What you can find:

  • Presale reports: Published 1-2 weeks before deal close, containing detailed pool statistics, structure analysis, and rating rationale
  • Surveillance reports: Ongoing monitoring of rated deals
  • Default studies: Historical default rates by rating category and asset class
  • Transition matrices: How ratings migrate over time
  • Criteria documents: Methodology explaining what drives ratings

Access:

  • Basic ratings: Free at spglobal.com/ratings
  • Full research: $25,000-100,000+/year depending on coverage
  • Some research available through Bloomberg

How to use it effectively:

  1. Read the Annual Default Study. It provides historical default rates by rating that you can use to benchmark your loss assumptions.
  2. Presale reports are dense with pool statistics. Even without a subscription, they sometimes appear in news or through banker contacts.
  3. Criteria documents are often free and explain exactly how S&P rates different asset classes. Read these before structuring a deal you want rated.

Moody’s

Similar to S&P: presale reports, surveillance, and default studies. Moody’s also publishes “idealized default rates” that are commonly used in CLO modeling.

Access:

  • Basic ratings: Free at moodys.com
  • Full research: Subscription required

Fitch ratings

Fitch is the third major agency and often rates deals that S&P and Moody’s decline (particularly in esoteric asset classes).

What makes it useful: If you are working on a less common asset class, Fitch may have relevant criteria and precedent deals when the others do not.

KBRA (Kroll bond rating agency)

KBRA has grown significantly in structured finance as an alternative to the big three.

What makes it useful:

  • Often rates deals the big three will not touch
  • More accessible research (some reports free after registration)
  • Timely commentary on market trends

Access: Register free at kbra.com for basic access

Using rating agency data effectively

Default studies: Compare your expected loss assumptions to historical data. If you are underwriting a BBB tranche with 100 bps expected annual loss and the rating agency shows BBB historically defaults at 200 bps over 5 years, check your math.

Transition matrices: Model what happens if ratings migrate. A AAA tranche that gets downgraded to AA may violate investor mandates, forcing selling pressure.

Presale reports: Free pool-level data on new rated deals. Even if you do not subscribe, these reports sometimes appear in news coverage or through your network.

Criteria documents: Understand what drives ratings before you structure a deal. If you know S&P requires 35% credit enhancement for AAA on your asset class, you can structure accordingly.

Choosing the right stack for your role

Do not over-subscribe. Start with free sources, add paid sources based on demonstrated need.

For originators seeking capital

Minimum viable stack (free):

  • SEC EDGAR: Understand how comparable deals are structured
  • Fed data: Track macro indicators affecting your asset class
  • One rating agency website: Basic ratings and free reports

Upgraded stack ($30,000-50,000/year):

  • Bloomberg terminal: Real-time market color, comparable deal tracking
  • Finsight: New issue flow and pricing

For capital providers building an ABF function

Minimum viable stack (~$50,000/year):

  • SEC EDGAR + TRACE (free)
  • Bloomberg terminal ($25,000): Market data and screening
  • One rating agency subscription ($25,000): Presales and surveillance

Full stack ($150,000-300,000+/year):

  • Intex: Cash flow modeling
  • Asset class-specific: dv01 (consumer), Trepp (CRE), LCD (leveraged loans)
  • Multiple rating agencies

For allocators and LPs

Minimum viable stack:

  • Manager reporting from your GPs (free)
  • One rating agency subscription: Validate manager claims
  • Bloomberg: Market context

Upgraded stack:

  • Direct data access (Intex, dv01) to spot-check manager portfolios
  • Industry databases (Preqin, PitchBook) for manager benchmarking

Practical tips for data extraction

Working with EDGAR

  1. Full-text search is your friend. Search by issuer name, deal name, or specific terms to find relevant filings quickly.

  2. ABS-EE parsing example (Python):

import pandas as pd
from lxml import etree

# Parse ABS-EE XML file
tree = etree.parse('abs_ee_file.xml')
root = tree.getroot()

# Extract loan-level data (structure varies by filing)
loans = []
for asset in root.findall('.//asset'):
    loan_data = {
        'loan_id': asset.find('loanId').text,
        'balance': float(asset.find('balance').text),
        'rate': float(asset.find('interestRate').text)
    }
    loans.append(loan_data)

df = pd.DataFrame(loans)

  1. Track 10-D reports. Set calendar reminders for monthly filings on benchmark deals. Servicer reports tell you how pools are actually performing versus projections.

  2. Read prospectuses for waterfall mechanics. The legal language is dense, but it tells you exactly how cash flows work.

Working with Bloomberg

  1. SRCH ABS <GO>: Filter deals by issuer, asset class, rating, vintage. Save searches for regular monitoring.
  2. DES <GO>: Enter any CUSIP to see deal description, pool statistics, and structure.
  3. YAS <GO>: Yield analysis, useful for relative value.
  4. MSG <GO>: Contact Bloomberg help desk. They know structured products well.

Working with rating agencies

  1. Register for free accounts. You get basic ratings and some reports without a subscription.
  2. Negotiate if you are a frequent user. Issuers and large investors often get discounted or complimentary access.
  3. Read criteria documents. These explain the methodology and are often free.

Building your own database

Start simple and expand based on need:

  1. Begin with a spreadsheet. Track deals you care about with key fields: issuer, deal name, close date, asset class, AAA spread, advance rate, source.

  2. Update monthly. Pull new issue data from Finsight, Bloomberg, or rating agency announcements.

  3. Migrate to a database. When your spreadsheet hits 500+ deals, move to PostgreSQL, Airtable, or similar. Structure matters more as data grows.

  4. Automate where possible. Set up APIs to pull FRED data automatically. Schedule reminders for manual updates.

Building a market intelligence workflow

Data is only useful if you actually look at it. Build habits around regular review.

Weekly routine

  • Monday: Review new issue calendar. What deals are coming this week? Any in your target asset classes?
  • Midweek: Check secondary trading levels (TRACE, Bloomberg) for deals you own or are tracking.
  • Friday: Read surveillance reports on deals you follow. Any watchlist additions or rating actions?

Monthly routine

  • Update your spread comps table with new deal pricing
  • Track 10-D filings for benchmark deals in your asset class
  • Review Fed data releases (G.19, charge-off rates)

Quarterly routine

  • Read rating agency transition studies and update default assumptions
  • Update your market overview deck for internal stakeholders
  • Review FFIEC data for warehouse capacity trends

Staying current

  • Subscribe to newsletters: TreppWire, bank ABS research, rating agency alerts. These synthesize information you would otherwise miss.
  • Follow key analysts on LinkedIn. Many share useful commentary.
  • Attend 1-2 conferences annually. Market color from hallway conversations is not captured in any database.

Important: Free data has a lag. SEC filings are 15-45 days behind deal close, TRACE has reporting delays, and Fed data is quarterly. For real-time intelligence, you need paid sources or direct relationships with market participants.

Note: The most valuable data source is often your network. A 15-minute call with a broker or capital provider gives you market color that no database captures. Use data to prepare for those conversations, not replace them.

← All Topics