Explore the critical theoretical underpinnings of the Black–Scholes–Merton model, its key assumptions, and how real-world markets often deviate from this foundational framework for pricing options.
The Black–Scholes–Merton (BSM) model changed everything in the options world. Before the model’s publication in the early 1970s, practitioners relied on a patchwork of trial-and-error, intuition, and partial equilibrium frameworks for pricing. But eventually, Fischer Black, Myron Scholes, and Robert Merton demonstrated how to systematically—and mathematically—value options under certain simplifying assumptions. Now, if you’ve ever traded options or even just thought about using them in a portfolio, you’ve probably encountered the BSM formula or some version of it.
In a nutshell, the BSM model provides a theoretical fair value for a European-style option written on a non-dividend-paying stock. Its derivation uses a clever “risk-neutral” approach, a bunch of calculus, and (surprise, surprise) a set of critical assumptions. Understanding those assumptions is a big deal. Even at a capstone CFA level, we still come back to the question: “How do these assumptions shape real-world pricing, and where do real markets deviate?”
In this section, we’ll run through the BSM assumptions, why they matter, and how acknowledging them can sharpen your sense of model applicability, hedging approaches, and exam readiness. The BSM model isn’t perfect. It can’t be, because markets are complicated beasts. But as you’ll see, it remains the cornerstone for more advanced frameworks that handle everything from stochastic volatility to jump processes.
The first assumption basically says, “We can trade any time with no transaction costs and no bid-ask spreads.” In other words, we can continuously rebalance our hedges and replicating portfolios without incurring fees or taxes.
• No transaction costs: This means no brokerage fee, no market impact, and no friction associated with each buy or sell.
• No bid-ask spreads: You always trade at the “fair price,” so to speak.
• Unlimited continuous trading: You’re allowed to trade at all times—think high-frequency trading on perpetual autopilot, but with no slippage or liquidity constraints.
Obviously, real markets have evolved since 1973, and transaction costs have come down a lot. Even so, they’re not exactly zero. Back in my early days (ah, the late 1990s), I once tried “rapid-fire” rebalancing in a small options account—just for the thrill of it—and found that commissions ate a surprising chunk of my gains. No model had warned me properly. So keep that in mind.
Next, the BSM model demands that the underlying asset price (here, the stock) follows a geometric Brownian motion (GBM) with a constant drift and volatility. In more technical terms, we assume:
(1) The stock price St evolves according to
dSt = μ St dt + σ St dWt,
where μ is the drift (expected return of the stock), σ is the volatility, and Wt is a standard Brownian motion.
(2) Log-returns (ln(St+1 / St)) are normally distributed.
(3) The volatility parameter (σ) is constant over the option’s entire life.
This construction is immensely powerful because it produces lognormal price distributions. But does real life always match a lognormal path with constant σ? Probably not—“volatility clustering” and “jumps” (like on major news events) are well-documented. That’s why you sometimes see skew or smile patterns in implied vol surfaces.
Another assumption is that the risk-free rate r is constant over the life of the option. The BSM formula essentially uses e^(-rT) as the discount factor for the final payoff, which implies you can either borrow or lend at a stable, unvarying interest rate. We pretend that short-term T-bill yields, repo rates, or central bank rates never shift for the duration.
In reality, the term structure of interest rates changes daily—sometimes intraday—and this can matter for longer-dated options. From a theoretical standpoint, if rates vary, you can still calibrate or adapt the model with a time-dependent discount function. But for a baseline exam question, you typically assume a single “flat” risk-free rate.
In its simplest form, the BSM model is built for European calls and puts on a stock that does not pay dividends. Once a stock pays a dividend, that changes the expected path of the underlying. The share price typically drops by the amount of the dividend on ex-dividend dates, so an unadjusted BSM formula would misprice calls (and puts) if it ignores that.
Of course, there is a well-known adjustment for discrete dividends (or a continuous dividend yield, q). Many real-world references give you the BSM formula adjusted for q, basically discounting the stock price at rate q. But if a question states “no dividends,” you can forget about ex-dividend date complexities.
Ever tried to borrow a large sum in real life? Yeah, banks don’t always offer unlimited lines of credit, especially not at a tiny risk-free rate. But academically, the BSM model demands that we can borrow (and lend) unlimited amounts of capital at the risk-free rate.
This assumption is essential to constructing the “perfect hedge” or replicating portfolio that underpins the option’s fair value argument. If you can’t borrow or lend freely, or your interest rate is higher than the risk-free rate, then your replication strategy might not fully hold.
Short selling the underlying asset is crucial for building the replicating portfolio for certain option positions. In the BSM framework, you can short shares without constraints (no margin requirements, no locate demands, no borrow costs).
That’s obviously not how modern capital markets operate. Short locating can be tricky, fees can be high (particularly in illiquid or “hard-to-borrow” stocks), and margin regulation definitely places constraints. Nonetheless, for theoretical replication, we suppose none of these issues exists.
In advanced option pricing, we rely on a principle called “no-arbitrage.” Essentially, if you can replicate the payoff of an option by trading the underlying asset and risk-free bonds, then the value of that option must match the cost of creating its replicating portfolio—otherwise, you can do an arbitrage. The BSM formula emerges from systematically applying that logic, but only after the above assumptions are baked into the mathematics.
The frictionless market assumption is key because if you had transaction costs, that’d erode your profits (or add cost) each time you rebalance. The entire replication argument rests on the idea that you can perfectly hedge an infinitesimal delta in the option with an offsetting position in the underlying—rebalanced continuously over the option’s life. Incidentally, “continuous rebalancing” is a theoretical ideal. In real markets, you typically can’t rebalance absolutely continuously, but you do it “frequently enough” for the strategy to approximate the model.
The assumption that volatility is constant (a direct outgrowth of the GBM approach) is the main reason the implied volatility surface is “flat” in the simplest BSM world. But in the real market, implied vol often differs by strike (smile or skew) and time to maturity (term structure). That’s partly a reflection that the BSM model’s constant volatility assumption doesn’t describe actual price dynamics of the underlying.
Given that actual markets violate many of the BSM assumptions, especially constant σ and frictionless trading, you often see price behaviors that the standard BSM formula can’t fully explain:
• Volatility Smiles and Skews. Empirical evidence shows that out-of-the-money puts on equity indexes tend to reflect higher implied volatilities, indicating bigger perceived tail risk than the lognormal assumption.
• Transaction Costs and Bid-Ask Spreads. Traders might not be able to perfectly hedge or arbitrage small mispricings.
• Dividend Effects. Real underlying assets often pay dividends, which can reduce call option prices (since the stock’s forward price is effectively lower once you factor in the dividend payout).
• Interest Rate Fluctuations. Over multi-year horizons, ignoring changes in the risk-free rate can affect forward pricing and option valuations.
Despite these snags, the BSM model still dominates as the starting point. Whether you see a hedge fund or a high-level portfolio manager at a bank, they rarely throw BSM out altogether. They just calibrate or modify it to handle reality better—via local volatility functions, jump-diffusion models, or even advanced frameworks like Heston (stochastic volatility).
A few well-known variations (or “next-generation” models) address the main BSM oversights:
• Black’s Model. Handles interest rate options and can incorporate forward-based references.
• Dividend Yield Models. Adjust for continuous dividend yield or discrete dividends (like the adjustments used in equity index options).
• Stochastic Volatility Models (e.g., Heston). Remove the assumption that σ is constant by letting volatility itself follow a stochastic process.
• Jump-Diffusion Models (e.g., Merton’s Jump-Diffusion). Add jumps to the price process, capturing sudden, large moves.
• Local Volatility Models (Dupire). Assume volatility is a deterministic function of time and price, capturing implied volatility smiles.
For the CFA exam, you might only need to know that these extensions exist and that standard BSM is not set in stone when it comes to actual trading. However, exam questions often revolve around using the plain BSM formula, identifying assumptions, and explaining which assumption might break. So if you’re aiming for a top score, keep these variations in mind.
Below is a simplified flowchart illustrating the skeleton of the BSM approach—from the underlying stock’s geometric Brownian motion to the final option pricing output.
flowchart TB
A["Starting Point: <br/> Spot Price S(0)"] --> B["Grow stochastically <br/>following GBM"]
B --> C["Apply <br/>Risk-Neutral Valuation"]
C --> D["Discount at <br/>constant r"]
D --> E["Option Price <br/>via BSM"]
This diagram intentionally strips away real-world complexities (e.g., transaction cost frictions, short-selling constraints, etc.). But it offers a visual peek at how the model sees the world.
Even though the CFA exam typically doesn’t require coding, it’s helpful to see how a standard BSM call price might be computed. Below is a concise Python snippet using the common BSM formula:
1import math
2from math import log, sqrt, exp
3from scipy.stats import norm
4
5def bsm_call_price(S, K, r, sigma, T):
6 """
7 Computes the Black-Scholes-Merton price of a European call.
8 S: Current stock price
9 K: Strike price
10 r: Risk-free rate (constant)
11 sigma: Volatility (constant)
12 T: Time to maturity (in years)
13 """
14 d1 = (log(S/K) + (r + 0.5*sigma**2)*T) / (sigma*sqrt(T))
15 d2 = d1 - sigma*sqrt(T)
16 call_price = S*norm.cdf(d1) - K*exp(-r*T)*norm.cdf(d2)
17 return call_price
18
19if __name__ == "__main__":
20 call_val = bsm_call_price(S=100, K=100, r=0.05, sigma=0.2, T=1)
21 print("Example BSM Call Price:", round(call_val, 2))
We see several assumptions right there: a constant r, constant σ, and no mention of dividends. If you run it, you’ll get a single “fair value” for the call. In real life, implied volatility can vary with strikes and maturities, so you might “back-solve” for sigma from the option’s market price. That’s the famous “implied volatility” concept.
From a regulatory standpoint, IFRS (International Financial Reporting Standards) or US GAAP typically require that derivatives (including many types of options) be reported at fair value on the balance sheet. In IFRS 2, for instance, share-based payments might be measured using option pricing models—often a BSM variant. Organizations subject to these standards need a recognized, systematic approach to valuation, so BSM serves as a standard reference point.
Additionally, the CFA Institute Code and Standards encourages “adequate diligence” in selecting and applying valuation methods, emphasizing the importance of clearly explaining assumptions to clients or stakeholders. Using a BSM approach without acknowledging potential limitations could lead to misaligned expectations. So best practice is to disclose model weaknesses—like ignoring dividends or transaction costs—and show possible sensitivity analyses.
• Mixing Up Variables: Confusing discrete compounding with the continuous discount factor e^(-rT).
• Forgetting Dividends: If a question states a dividend yield, you can’t just apply the standard BSM formula.
• Overcomplicating “Constant” Volatility: The exam might simply say “sigma is 20%”—just proceed; no need to tweak or guess volatility skew.
• Using American vs. European. BSM is strictly for European-style options. If the question is about American-style, watch out for potential early exercise (especially with dividends).
• Sign Errors in d1 and d2: The formula’s sign matters. A small mistake can lead you to the wrong final number.
The Black–Scholes–Merton model is a timeless beacon in the world of option pricing. Knowing its assumptions isn’t just about memorizing bullet points for the exam—it’s understanding why the model works in a frictionless, continuous environment and how real markets inevitably deviate. By studying these assumptions, you’ll gain an appreciation for both the elegance and the limitations of the BSM formula. Yes, real markets have complexities like discrete dividends, varying interest rates, and transaction costs, but BSM remains the foundation upon which advanced models are built.
In your career (and especially as you prepare for the CFA exams), you’ll likely reference BSM or an adapted form at some point. Experience from the field—trust me on this—teaches you that no model is perfect. But BSM still gives a baseline for pricing and a shared language for discussing volatility, risk, and payoff structures. As you move on, keep an eye open for ways that real world differs from theoretical markets, but also appreciate the mathematical clarity that Black, Scholes, and Merton brought to finance.
• Black, Fischer, and Scholes, Myron. “The Pricing of Options and Corporate Liabilities.” Journal of Political Economy, 1973.
• Merton, Robert C. “Theory of Rational Option Pricing.” The Bell Journal of Economics, 1973.
• Hull, John C. “Options, Futures, and Other Derivatives.”
• IFRS 2: Share-Based Payment.
• CFA Institute Program Curriculum: Derivatives and Risk Management.
Important Notice: FinancialAnalystGuide.com provides supplemental CFA study materials, including mock exams, sample exam questions, and other practice resources to aid your exam preparation. These resources are not affiliated with or endorsed by the CFA Institute. CFA® and Chartered Financial Analyst® are registered trademarks owned exclusively by CFA Institute. Our content is independent, and we do not guarantee exam success. CFA Institute does not endorse, promote, or warrant the accuracy or quality of our products.