The First Decision: Immediacy or Price Control?
For a retail equity order, the first consequential decision is not which venue appears in a routing disclosure. It is whether the investor wants the market to provide a trade now or is willing to wait for a chosen price. This chapter isolates that choice so readers can classify marketable and nonmarketable limit orders, see what the bid-ask spread pays for, and judge when a limit order’s apparent price advantage is actually an exchange of certainty for uncertainty.
The distinction is practical. Someone trying to exit a position before a deadline has a different problem from someone willing to buy only at a specified price. The two investors may use the same commission-free account and trade the same stock, yet their order instructions express different priorities. Execution quality cannot be assessed fairly until those priorities are visible.
What the order says about the prevailing market
A marketable order is executable when the broker receives it. The category includes ordinary market orders and limit orders whose prices are aggressive enough to trade against the opposite side of the prevailing National Best Bid and Offer, or NBBO. For a buy order, that generally means the limit price is at least as high as the best offer; for a sell order, it is at least as low as the best bid. The exact order instruction may differ, but the economic intention at that moment is similar: obtain an execution from the liquidity currently available.
A nonmarketable limit order does something else. It specifies a price without crossing the opposite quote at arrival. A buy limit below the best offer, or a sell limit above the best bid, waits for the market to move toward the investor’s price. The order does not ask the broker to accept any price available now. It asks the market to meet a boundary set by the investor.
An illustration makes the distinction concrete. Suppose a stock is quoted at a $49.99 bid and a $50.01 offer. A marketable buy order asks for an immediate purchase at the available offer or at another immediately executable price. A buy limit at $49.99 does not accept the $50.01 offer. It is willing to buy at $49.99 or better, but it may wait if sellers do not reach that price. The $0.02 gap is the quoted spread in this simplified example. It is not a commission, but it is part of the price of obtaining liquidity immediately.
The spread is best understood as a compensation mechanism, not a moral penalty. A party willing to sell now must be ready to meet a buyer’s demand, and a party willing to buy now must be ready to meet a seller’s demand. The marketable order receives immediacy by taking liquidity that another participant has made available. The cost of that immediacy can appear as a less favorable execution relative to the midpoint between the bid and offer. A limit order attempts to avoid paying that full immediacy cost by offering liquidity on its own terms.
That description does not mean a marketable order is always the wrong choice. It can be appropriate when the value of immediate completion exceeds the potential saving from waiting. Nor does a limit order always receive a better result. Its price boundary can protect the investor from paying more than intended, but the boundary can also prevent the trade from happening.
The order type therefore changes what is certain and what is not:
- With a marketable order, execution is the immediate objective. The investor gives up some control over the exact price in exchange for a high probability of completion and a short wait.
- With a nonmarketable limit order, the price boundary is more certain, but the time and completion of the trade are not. The investor may obtain a favorable price on filled shares, or may receive only part of the intended position, or none of it.
The terms “marketable” and “nonmarketable” should not be mistaken for judgments about sophistication or quality. They describe the relationship between the order’s price and the prevailing quote at the time the broker receives it. A limit order can be marketable. A marketable limit order is still a limit order in its written instruction, but it is executable immediately and belongs with marketable orders for the study’s comparison. A nonmarketable limit order is the resting, price-controlled case that carries the distinctive risk of non-execution.
The diagram’s central point is not that one branch is superior. It is that the branches move uncertainty to different parts of the outcome. The marketable branch makes waiting less uncertain but makes the execution price less controlled. The nonmarketable branch makes the price boundary more controlled but makes completion dependent on what the market does next.
When price control is worth the wait
The economic appeal of a nonmarketable limit order depends first on the investor’s urgency. Suppose the order is part of a hedge, a required rebalancing, or an exit before a known deadline. In those situations, an unfilled order can be more costly than paying the spread. A price boundary that looks prudent in isolation may fail the investor’s actual objective if it leaves the necessary position unbuilt or the unwanted position open.
Now suppose the investor is making a discretionary purchase and is willing to miss it rather than pay above a chosen level. The same non-execution is not necessarily a failure. The order did what its instruction required: it refused to trade beyond the limit. The economic judgment then turns on whether the investor valued the opportunity to own the stock more than the price concession needed to obtain it. Execution quality is not just a property of the broker; it is partly a fit between the order instruction and the objective behind it.
The quoted spread changes that calculation. When the spread is narrow, the price concession for immediacy may be small, so waiting may save little while still creating a chance of non-execution. When the spread is wide, the potential compensation for providing liquidity is larger. A patient limit order has more room to improve on the midpoint comparison if it fills, although its probability of filling and the cost of waiting still matter.
The retail evidence is consistent with that conditional logic. In the study’s sample, the estimated cost advantage of limit orders is larger in smaller stocks, where quoted spreads are wider. The paper reports an approximately 20-basis-point difference between limit and marketable orders in smaller stocks, compared with a smaller difference in large stocks. This is not a promise that every small stock rewards patience. It shows why the market’s compensation for immediacy is one of the conditions that can change the relative appeal of the order types.
Volatility complicates the picture. A moving market can bring a patient limit price into reach, increasing the chance of execution. But the same movement can carry the market away from the investor’s price, leaving an unfilled order behind. The study finds that the relative advantage of limit orders grows when volatility is higher, but that result should not be read as “volatility makes limit orders safe.” It means that, in the measured sample and under the study’s implementation-shortfall framework, the comparison shifted toward limit orders as the market became more variable. The risk of adverse price drift remains part of the mechanism.
Liquidity matters for a related reason. A small, actively traded stock may offer a narrow spread and frequent opportunities to trade. A less liquid stock may offer a wider spread and a larger apparent reward for supplying liquidity, but fewer participants willing to meet the limit price. The investor is balancing a larger possible saving against a less dependable path to completion. The right question is not simply whether the spread is wide; it is why it is wide and whether the investor can tolerate the order’s remaining open.
Price control is also a changing state, not a permanent label attached to the order. A buy limit that begins below the offer can become executable if the market moves down to the investor’s price. At that point, the order has crossed an important boundary: it is no longer merely waiting for a possibility; it is eligible to trade against the liquidity then available. If it fills, the investor has converted patience into a controlled execution. If it remains open or is canceled before that point, the price boundary has protected the investor but has not produced the intended position.
That sequence explains why the same limit price can have different meanings on different days. In a quiet market, a small distance from the quote may be reached quickly. In a fast market, the quote may move through the price and then continue away, leaving the order unfilled even though the market briefly touched a favorable level. A wide spread can increase the potential saving from waiting, but it can also signal that fewer willing counterparties are available. Volatility can create more chances to reach the limit and more chances for the opportunity to disappear. The order’s value is therefore produced by the interaction between its boundary and the path the market takes, not by the limit price alone.
Monitoring is another part of the price-control bargain. A limit order is not a one-time declaration that can be detached from changing conditions. If the market moves, the investor may need to decide whether the original price boundary still reflects the objective. Leaving an order unchanged can be a deliberate choice to preserve discipline. It can also become an unintended exposure if the investor has forgotten the order, if the reason for the trade has changed, or if the market has moved so far that completing the position later would be materially different.
This is why “use limit orders to save money” is too simple. A limit order can reduce the price paid for the shares that execute, but it may also delay or prevent the desired trade. A marketable order can cost more in spread terms but protect the investor from missing an urgent opportunity. The correct choice depends on which uncertainty the investor is willing to bear: uncertainty about price, or uncertainty about completion and timing.
A useful test is to ask three questions before treating patience as a saving. If the order fills later today, will the position still serve the same purpose? If it never fills, is missing the trade acceptable? If the price moves away, would completing the order later be worse than accepting the spread now? These questions do not predict the market. They identify which consequence the investor is actually trying to avoid.
For an urgent sale, the main risk may be that the position remains open while the price moves against the investor. For a discretionary purchase, the main risk may be paying more than the investor considers reasonable. In the first case, a nonmarketable limit order can fail through non-completion even if its price discipline is flawless. In the second, a marketable order can fail the investor’s stated price objective even if it executes immediately. The order instruction is effective only when it expresses the objective rather than replacing it with a generic preference for “better price” or “faster execution.”
There is one further distinction that becomes essential when these choices are compared later. The intention to complete the full order cannot be assumed silently. If an investor submits a limit order precisely because they are willing to miss the trade, then treating every unfilled share as though it had to be purchased immediately will overstate the order’s personal cost. If the investor was committed to obtaining the full position, then ignoring the unfilled shares will understate it. A sound comparison must say which intention it is measuring.
The evidence in the retail limit-order study adopts the second perspective for its implementation-shortfall analysis: it treats the trader as committed to filling the full top order and imputes a cost for the unfilled portion. That choice makes marketable and nonmarketable orders comparable as ways of completing an intended trade, but it does not describe every investor’s preference. It is a measurement assumption that exposes an important economic consequence, not a rule that every unfilled limit order was a mistake.
The central decision can now be stated without a universal prescription. Choose a marketable order when completion and timing are worth accepting the prevailing liquidity cost. Choose a nonmarketable limit order when the price boundary matters more than immediate completion and the investor can accept waiting or missing the trade. Spread width, volatility, liquidity, and monitoring capacity can tilt the calculation, but none removes the underlying exchange.
The remaining question is how to count the outcome when the trade fills partly, fills later, or does not fill at all. Execution price alone cannot answer that question. The next chapter turns to implementation shortfall, which combines the cost of executed shares with the opportunity cost assigned to shares left unexecuted.