What it means
Reliable historical data is the raw material of financial research. CRSP collects prices, dividends, trading volume and other details for a very large number of listed securities and organises them in a consistent way.
This helps researchers avoid errors and biases that arise when data is gathered from many different sources. One key feature is that the database includes companies that no longer exist.
Studies that look only at today's surviving companies suffer from survivorship bias, meaning they overstate performance because the failures have disappeared from the data. By keeping delisted companies in the record, CRSP gives a more honest picture.
The database also adjusts for events such as share splits and dividends so that returns can be calculated accurately across time. A researcher can see the total return, which combines the price change and any income paid to shareholders, rather than just the movement in the share price.
CRSP is widely used by universities, central banks, asset managers and consultants. Many famous studies on topics like the size effect (smaller companies sometimes outperform) and momentum (recent winners often keep winning) have relied on it or similar data.
Access usually requires a subscription, often through a university or a data provider, so individuals rarely use it directly. Non-specialists are more likely to meet it indirectly, through research reports or market indexes that cite its results.
The data is a record of the past, so it cannot predict the future. Researchers treat it as evidence to be tested and questioned, not as a guarantee.
In practice
Real-world examples.
Example
A finance professor studies whether small company shares outperform large ones over fifty years. She uses CRSP data because it covers delisted companies and includes dividends. Her results can be repeated by other researchers, because they can use the same database and check the method.
Example
An asset manager tests a new investment strategy on historical data before launching a fund. The team uses CRSP to avoid survivorship bias that would flatter the results. If the strategy only works on current survivors and fails on the full data, the team knows to discard it before clients' money is at risk.
Example
A central bank economist examines how stock market returns relate to economic cycles. The long, consistent record lets her compare different decades on the same basis. Her findings help the bank judge how stock markets may respond when the economy slows or recovers.
Formula
Calculation
Total return = (Ending price - Beginning price + Dividends) / Beginning price x 100
Worked example: a share is priced at $50 at the start of the year and $55 at the end. It pays dividends of $2 per share during the year.
Price gain = $55 - $50 = $5.
Total gain including dividends = $5 + $2 = $7.
Total return = $7 / $50 x 100 = 14%.
A database like CRSP records both the price and the dividend, so that this total return can be calculated consistently for every security and period.Case study
Seen in the real world.
Delmar Analytics is a fictional research firm, and this case is illustrative only. It built a strategy that bought shares with strong recent returns, and its early tests on a list of current companies looked excellent.
A senior analyst pointed out that the test excluded companies that had failed or been delisted. When the firm repeated the work using a database that included delisted companies, the strategy's returns fell by about a third.
The firm adjusted its claims and explained the results more cautiously to clients. The analyst said the experience showed why complete historical data is not a luxury but a basic requirement. The experience also changed the firm's review process, so every backtest now has to state clearly what data it used and whether failures were included.
Watch out
Common mistakes.
- Using only today's surviving companies in a historical test, which overstates performance. Always check whether a data source includes companies that have been delisted or merged.
- Ignoring dividends and looking only at price changes, which understates total returns. Total return is the fairer measure because shareholders receive both price gains and dividends.
- Treating past results as a promise about future returns. History is a guide to how markets have behaved, not a promise of what will happen next.
Questions
People also ask.
What does CRSP stand for?
It stands for the Centre for Research in Security Prices, a research group at the University of Chicago Booth School of Business. It is based at a university and is best known for its long historical record of US stock prices and returns.
Who uses CRSP data?
University researchers, central banks, asset managers and consultants studying market behaviour. Many well-known academic studies of market behaviour have been based on this kind of data.
Why does survivorship bias matter?
Because excluding failed companies makes past returns look better than they really were, which can lead to poor investment decisions. Complete data helps analysts and investors avoid being misled by a record that only includes the winners.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%