We must decide whether the district court erred in upwardly adjusting the base offense level under the federal Sentencing Guidelines for abuse of a position of public or private trust.

I

On March 21 and 22,1988, in his capacity as an employee-driver for Interstate Container, Inc. of Long Beach, California (“Interstate Container”), Hill picked up in Kansas and Missouri furniture and several crates containing the household goods and personal possessions of five military families (“families”) who were being relocated to Germany. Hill was responsible for transporting their belongings to Texas, where they would eventually be shipped abroad.

On March 26, 1988, instead of delivering the belongings to Texas, Hill and his brother-in-law opened the crates and sold or traded several of the household items to various individuals in and around Licking, Missouri. Some of the families’ personal belongings, including photographs, cards, and letters, were destroyed. In exchange for one of the stolen items, Hill obtained the use of a barn in Licking in which he stored some of the misappropriated crates.

Hill was apprehended, and on April 11, 1988, a federal grand jury in the Western District of Missouri returned a three-count indictment against him. Hill was charged with one count of conspiracy to commit theft in violation of 18 U.S.C. §§ 371 and 659 and two counts of theft of an interstate shipment in violation of 18 U.S.C. §§ 2 and 659. Hill consented to transfer of the case to the Central District of California.

On November 21, 1988, Hill pleaded guilty to one count of conspiracy to commit theft of an interstate shipment. On January 23, 1989, over Hill’s objection, the district court followed the recommendation of the pre-sentence report and adjusted Hill’s base offense level upward under section 3B1.3 of the Sentencing Guidelines for abuse of a position of trust. It then sentenced the defendant to five months’ imprisonment, five months in a treatment center, and three years of supervised release. The court also ordered Hill to pay restitution. The remaining two counts were dismissed.

Hill timely appeals from the district court’s sentence; he was released on bond pending this appeal. We have jurisdiction under 18 U.S.C. § 3742 and 28 U.S.C. § 1291.

II

Section 3B1.3 of the Sentencing Guidelines mandates a two-level upward adjustment of a defendant’s base offense level “[i]f the defendant abused a position of public or private trust ... in a manner that significantly facilitated the commission or concealment of the offense.” United States Sentencing Commission, Guidelines Manual § 3B1.3, at 3.7 (Nov.1989). Hill contends that application of this section was improper because the relationship between a truck driver and the owner of the truck’s cargo does not give rise to a position of trust. Hill further argues that he was not in a position of trust vis-a-vis the families because he was merely an employee-driver of Interstate Container and[*505] had not been personally sought out by the families.

What is the appropriate standard of review? It is axiomatic that legal issues are reviewed de novo and factual questions are reviewed for clear error. See, e.g., United States v. Foreman, 905 F.2d 1335, 1338 (9th Cir.1990). We have consistently held that a district court’s application of the guidelines is reviewed de novo. See, e.g., United States v. Watt, 910 F.2d 587, 589 (9th Cir.1990); United States v. Heldberg, 907 F.2d 91, 93 (9th Cir.1990). These principles, however, do not necessarily resolve the issue; whether a district court has made a factual determination, a legal conclusion, or both, can sometimes be difficult to discern in a Sentencing Guidelines case. Cf. United States v. Gonzalez, 897 F.2d 1018, 1019 (9th Cir.1990) (downward departure for acceptance of responsibility under section 3E1.1 reviewed only for clear error); United States v. Christman, 894 F.2d 339, 342 (9th Cir.1990) (upward adjustment for obstruction of justice under section 3C1.1 reviewable for clear error); United States v. Sanchez-Lopez, 879 F.2d 541, 557 (9th Cir.1989) (downward adjustment for minimal participation under section 3B1.2 reviewed for clear error); United States v. Mejia-Orosco, 867 F.2d 216, 221 (5th Cir.), cert. denied, — U.S. —, 109 S.Ct. 3257, 106 L.Ed.2d 602 (1989) (rationale for application of clearly erroneous standard is that these sections “call[ ] upon district judges to make sophisticated factual determinations” in individual cases).

In Foreman, we held that “whether an abuse of a position of trust must implicate a special privilege accorded someone in that position” was a legal issue, reviewable de novo. See Foreman, 905 F.2d at 1338.

The reasoning of Foreman compels the conclusion that de novo review should be applied here. The question of whether a truck driver may be in a position of trust vis-a-vis those individuals who own the truck’s cargo is a legal issue, much as the “special privilege” question in Foreman was also a legal issue. Thus, we need only consider whether the district court construed the Guidelines correctly.

Ill

Turning to the merits, Hill argues that truck driving is not the type of employment that can ever give rise to a position of public or private trust. We disagree.

A

The Commentary to the Guidelines Manual, Section 3B1.3 Application Note 1, provides that:

The position of trust must have contributed in some substantial way to facilitating the crime and not merely have provided an opportunity that could as easily have been afforded to other persons. This adjustment, for example, would not apply to an embezzlement by an ordinary bank teller.

It would appear that two characteristics of the position support the conclusion that an ordinary bank teller is not in a position of trust. ^[1] First, there is a simple, objective method of determining whether a teller has embezzled any funds which pass through the teller’s hands: if the bank till does not balance at the end of the teller’s shift, the teller is presumed to be responsible for the missing funds. Second, surveillance is easy; the teller is normally in plain view. The branch manager need only look up from her or his desk to determine whether all tellers are at their appropriate stations; the absence of a teller from his or her window would be quickly noticed.

[*506] In United States v. Ehrlich, 902 F.2d 327 (5th Cir.1990), pet’n for cert. filed (U.S. Sept. 12, 1990) (No. 90-5728), the Fifth Circuit concluded that a bank loan clerk was also in a position of trust. Id. at 330-31. The defendant used her position at MedCentre Bank to transfer funds from MedCentre's general ledger account to her own checking account. “Ehrlich’s position gave her the authority to routinely initiate loan balancing transactions ... [and] to balance the loan suspense account.” Id. at 331. As a result of her authority, Ehrlich’s embezzlement could easily escape the attention of her superiors at the bank. Id.; see also United States v. McMillen, 917 F.2d 773, — (3d Cir.1990) (savings and loan branch manager occupied position of private trust).

A simple principle emerges from these assorted authorities: the primary trait that distinguishes a person in a position of trust from one who is not is the extent to which the position provides the freedom to commit a difficult-to-detect wrong. If a person is in a relationship such that any attempt by a defendant to abuse the relationship could be simply or readily noticed by the second party to the relationship, presumably the two persons have not formed a “trust” relationship. Conversely, if one party is able to take criminal advantage of the relationship without fear of ready or quick notice by the second party, the second party has clearly placed a level of trust in the first.

These authorities also reveal two indicia of a position of trust. One indicium of such freedom to commit a difficult-to-detect wrong under section 3B1.3 is the inability of the trustor ^[2] objectively and expediently to determine the trustee’s honesty. A criminal act which cannot be discovered as a matter of routine is such a “difficult-to-detect” wrong. In contrast, the daily audit of an ordinary bank teller’s till is an “objective and expedient” means of discovering criminal activity, which supports the conclusion that a teller is not typically in a position of trust.

A second indicium of freedom to commit a difficult-to-detect wrong is the ease with which the trustee’s activities can be observed. This indicium implicates a defendant’s ability to make an undetected post-crime flight. If one has placed a defendant in a position to commit a crime and leave the area before the crime will be discovered, the defendant will generally be in a position of trust. ^[3]

B

With this principle and its corresponding indicia in mind, we turn to Hill. At first blush, the objective determination of honesty indicium of a position of trust suggests that there was not a trust relationship between the families and Hill. The families had a simple, expedient method by which to determine whether Hill had criminally disposed of their property: they simply had to compare the contents of the truck at its destination with the contents of the truck as it left their homes. If the truck arrived in Texas sans various items, the families could presume the items had been stolen. Moreover, absent evidence of forced entry into the truck, the families could presume that Hill was the culprit.

Under the unique circumstances of this case, however, this factor weighs in favor[*507] of a trust relationship. Because the families were relocating to Europe, the Texas destination point was only an intermediate stop for the belongings. Accordingly, the families had an objective means of determining Hill’s honesty, but not an expedient one.

The second indicium clearly favors the finding of a position of trust. Hill’s activities as a long-distance truck driver, almost by definition, were difficult, if not impossible, to observe during his cross-country trek. Hill had the opportunity — of which he took advantage — to abscond with the truck’s contents and to have long disposed of the goods before the owners were even aware of the goods’ disappearance.

Although neither of these indicia control, we find on balance that the district court did not err in finding that Hill was in a position of trust vis-a-vis the families. Hill had unwatched and exclusive control over the families’ belongings for an extended period of time. Moreover, the families did not have the ability to monitor Hill’s integrity because of their relocation to Europe.

Finally, Hill argues that he did not occupy a position of trust because the families did not seek him out personally. In essence, Hill argues that the families placed their trust in Interstate Container, Hill’s employer, and not Hill himself. We agree with Hill that, in many instances, a customer of an organization places her or his trust in the organization, and not in the individual server. However, such is not the case here. When Hill arrived at the families’ doorsteps, the families knew that Hill would be their exclusive server. They knew that Interstate Container would not and could not monitor Hill’s whereabouts for the duration of the trip. Accordingly, the families’ trust was placed as much in Hill as an individual as it was in the business that employed Hill.

IV

Hill’s having occupied a position of trust, it must next be determined whether he abused this position “in a manner that significantly facilitated the commission or concealment of the offense.” U.S. S.G. § 3B1.3, at 3.7. An upward adjustment is not appropriate if the position of trust “merely ... provided an opportunity that could as easily have been afforded to other persons.” Id. Hill argues that his position “provided him with an opportunity that could have been afforded to any other truck driver,” and thus, application of the upward adjustment was inappropriate. Appellant’s Opening Brief, at 11. We disagree.

By characterizing the argument in terms of an opportunity afforded “to any other truck driver” rather than “to other persons,” Hill begs the very question to be answered. By this logic, the defendant in United States v. Parker, 903 F.2d 91 (2d Cir.), cert. denied, — U.S. —, 111 S.Ct. 196, — L.Ed.2d — (1990), a security guard ultimately convicted of, inter alia, conspiracy to commit and commission of a robbery of an armored car, could have argued that “any other security guard” would have had the same opportunity to provide inside information to ccvconspira-tors. See id. at 104. ^[4] Likewise, Ehrlich could claim that “any bank loan clerk” would have the same opportunity to embezzle. See Ehrlich, 902 F.2d at 330-31. Finally, the defendant in United States v. Zamarripa, 905 F.2d 337, 340 (10th Cir.1990), could argue that any babysitter, not merely himself, would have been afforded the same opportunity to commit a sexual crime. This reasoning is incorrect. Bather, the question is whether Hill, relative to[*508] all people in a position to conspire to steal goods from interstate commerce (i.e., the public at large), was in a superior position as a result of a trust relationship.

Hill clearly was in a superior position. He was invited into homes so that he might, without direct supervision, transport the families’ personal items elsewhere. As a result of this trust, Hill had the ability to abscond with the belongings, move them to a new location, and dispose of them, all without fear of timely detection.

AFFIRMED.