A similar strategy might be used by the private buyer of a right-of-way even if it is known that a highway is going to be constructed. Suppose, for example, that the highway provider chooses more than one potential route. He then informs the land owners along the two routes that he would like to purchase specified parcels from each of them and that each should submit the price at which he is willing to sell (alternatively the developer might make initial bids that can be accepted, but indicate that each seller has an opportunity to make one take-it-or-leave-it counteroffer). In addition, potential sellers are informed that the buyer will purchase only the right-of-way with the lower total cost (a maximum might also be specified or information provided about the total acreage required on each right-of-way). Pipeline builders, for example, although their precise strategies may differ, "routinely consider alternative routes, negotiate with different groups of owners, and settle with the first group that comes up with an acceptable arrangement. Where buyers compete with competing groups of sellers, there is extra pressure on the sellers to agree to reasonable deals" (Roth 1996, 199). When a private buyer structures the bargain appropriately (for example, by secretly buying land, or by simultaneously considering alternative routes and buying the parcels only after every seller has agreed and before the project starts, or by choosing routes where landowners give up only part of their land and expect to collect increased rent on the rest), holdouts are not likely to prevent right-of-way acquisition. Indeed, as Roth notes, the first two modern privately provided highways in the United States--the Dulles Greenway in Virginia and SR-91 in California--obtained the land they required from private landowners without relying on eminent domain, choosing instead to bargain (some properties were in existing public roadway corridors, which had to be obtained from governments) (1996, 199).