Appellant Carl Bradley Johansson (“Jo-hansson”) pled guilty to a conspiracy to violate federal regulations which limit the number of hours operators of motor carriers may drive, and to conceal such violations by falsifying records required to be maintained by the United States Department of Transportation (“DOT”). He appeals the district court’s enhancement of his sentence pursuant to U.S.S.G. § 2F1.1(b)(6)(A) for a fraud offense that involves the conscious or reckless risk of serious bodily injury. He contends (1) that the district court erred in applying a preponderance of the evidence standard to find the conduct supporting the enhancement, (2) that even if the preponderance of the evidence standard applied, the evidence did not support the enhancement, and (3) that under Apprendi v. New Jersey, 530 U.S. 466 , 120 S.Ct. 2348 , 147 L.Ed.2d 435 (2000), a jury was required to find beyond a reasonable doubt the facts supporting the enhancement. We have jurisdiction pursuant to 18 U.S.C. § 3742 (a) and 28 U.S.C. § 1291 , and we affirm.

Johansson was the president and owner of a privately-owned trucking business that operated under the names Petroleum Delivery Service, Ash Incorporated, Ash Transportation, Atlas Carriers, Inc., and Atlas Bulk, Inc. (collectively “Atlas”). Atlas specialized in transporting hazardous materials such as petroleum fuels, primarily gasoline, and had combined annual revenues approaching $100 million. Atlas maintained corporate offices in Paramount and Montebello, California, and operated truck terminals throughout California.

On March 5, 1999, the government charged Johansson in a three-count superseding information. Count Three charged Johansson with a conspiracy to violate DOT regulations governing the number of hours drivers transporting hazardous materials may drive and the maintenance of daily logs that record the hours worked. See 49 C.F.R. § 395.3 (hours requirements); 49 C.F.R. § 395.8 (b) (daily log requirement). In particular, Johansson was charged with a conspiracy to knowingly and willfully make false statements in matters within the jurisdiction of the Federal Highway Administration (“FHWA”) (a division of the DOT) in violation of 18 U.S.C. § 1001 , by causing Atlas drivers to violate the hours-of-driving regulations and to create and maintain false and fraudulent daily logs for inspection by FHWA inspectors in order to conceal the violations.

Pursuant to a plea agreement, Johans-son pled guilty to all three counts of the information. With respect to Count Three Johansson agreed to (1) a base offense level of six; (2) a two-level enhancement for more than minimal planning; and (3) a four-level increase for being a leader, organizer, manager, and supervisor in a crimi nal activity involving five or more participants. The government also agreed to recommend a three-level or two-level reduction (depending on the total offense level) for acceptance of responsibility. It reserved the right to argue at the time of sentencing that Johansson’s offense level should be increased pursuant to U.S.S.G. § 2F1.1(b)(6)(A) because his offense involved the conscious or reckless risk of serious bodily injury.

The Presentence Report recommended the enhancements stipulated to in the plea agreement, as well as the § 2F1.1(b)(6)(A) enhancement, and a three-level reduction for acceptance of responsibility, for a total of offense level of 14. Johansson fell within criminal history category I with a resulting Guideline range of 15 to 21 months.

The parties subsequently filed their respective sentencing position papers. Jo-hansson argued that the government could not prove the conscious or reckless risk of serious bodily injury enhancement with respect to any of the counts, and the government recommended an increase in offense level pursuant to § 2F1.1(b)(6)(A) for the falsification of the daily logs offense. In support of its recommendation, the government submitted reports of two interviews of Gregory Surprenant, Atlas’ former manager at its Montebello terminal. According to the reports, for about six to eight months prior to July 1996, Johansson directed Surprenant to pay drivers cash under the table for hours that they worked in violation of DOT regulations. The cash loads were documented on a cash reconciliation form and maintained by Johansson separately and apart from the legitimate driver logs. Surprenant stated that Jo-hansson was actively involved in and directed the hours-of-driving violations and “cash loads.” Surprenant also reportedly stated that Johansson directed him to “keep all of the paperwork connected to the cash loads away from the safety department [of Atlas].” Johansson responded by submitting a written statement from J.T. Gibbons (“Gibbons”), a transportation expert. Gibbons opined that the falsification of the logs books did not pose a conscious or reckless risk of serious bodily injury.

The trial court held a sentencing hearing on January 20, 2000. At the hearing defendant again objected to application of the § 2F1.1(b)(6)(A) enhancement. The court concluded that it could not automatically assume, based on Johansson’s guilty plea to making false statements relating to the transportation of hazardous material, that his offense created a conscious risk of serious bodily injury. The trial court also concluded, however, that under the relevant conduct provisions of the Guidelines the government might be able to persuade him that the enhancement applied if it could show that the hours-of-driving violations and falsification of the daily logs were a regular part of Johansson’s business. The court accordingly continued the sentencing to a later date to allow the parties to develop further evidence.

The district court held the continued sentencing hearing and heard live testimony on March 24, April 6, and April 7, 2000. The government called two FHWA Inspectors, Amy Hope and Donald Carr, as well as James Bolla, a former Atlas Safety Manager. Defendant called his expert, Gibbons, as well as Surprenant, the former Atlas terminal manager. At the conclusion! of the evidence the district court allowed the parties to, once again, argue their respective positions. The court, applying a preponderance of the evidence standard, found that Johansson was aware of the fraudulent record-keeping practices, and indeed, authorized the practices, and that he did so in order to satisfy business needs. The court also found that hours-of- driving violations by drivers carrying hazardous material creates a substantial risk to the safety of other drivers on the road, and that Johansson was aware of that safety risk but nonetheless allowed the hours-of-driving violations to continue “on a regular basis.” Accordingly, the court applied the § 2F1.1(b)(6)(A) enhancement to Jo-hansson’s sentence.

As a result of the court’s finding, Jo-hansson’s total offense level was 14 with a corresponding range of 15 to 21 months. If the district court had not applied the enhancement, Johansson would have had an offense level of 10 with a corresponding range of six to 12 months, and he also would have been eligible for a “split sentence” pursuant to U.S.S.G. § 501.1(c). The district court ultimately sentenced Jo-hansson to 15 months imprisonment. The issues on this appeal involve the district court’s finding that Johansson’s offense involved the conscious or reckless risk of injury.

Johansson first contends that the district court’s application of a preponderance of the evidence standard of proof, rather than a clear and convincing standard, to find that Johansson’s conduct involved the “conscious or reckless risk of serious bodily injury” violated due process because the resulting increase in his offense level had “an extremely disproportionate effect on his sentence relative to the offense to which he pled guilty.” We review the constitutionality of a sentence de novo. See United States v. Mezas de Jesus, 217 F.3d 638, 642 (9th Cir.2000).

The Sentencing Guidelines comment “that it is ‘appropriate’ that facts relevant to sentencing be proved by a preponderance of the evidence.” United States v. Watts, 519 U.S. 148, 156 , 117 S.Ct. 633 , 136 L.Ed.2d 554 (1997) (citing U.S.S.G. § 6A1.3, Commentary). In United States v. Restrepo, 946 F.2d 654 (9th Cir.1991) (en banc), we held that the preponderance of the evidence standard of proof for Guidelines factors generally satisfies due process:

Id. at 659 (footnote omitted). We also noted, however, that “there may be an exception to the general rule that the preponderance standard satisfies due process when a sentencing factor has an extremely disproportionate effect on the sentence relative to the offense of conviction,” but that such an exception had nothing to do with the case. Id. at 659-60 ; see also Almendarez-Torres v. United States, 523 U.S. 224, 248 , 118 S.Ct. 1219 , 140 L.Ed.2d 350 (1998) (explicitly stating that it was not expressing a view as to whether a heightened standard of proof applies to sentencing determinations that bear significantly on the severity of sentence); Watts, 519 U.S. at 156-57 , 117 S.Ct. 633 (acknowledging that there is “a divergence of opinion among the Circuits as to whether, in extreme circumstances, relevant conduct that would dramatically increase the sentence must be based on clear and convincing evidence”).

Since Restrepo we have twice held that the preponderance standard did not satisfy due process. See United States v. Mezas de Jesus, 217 F.3d 638, 642-644 (9th Cir.2000); United States v. Hopper, 177 F.3d 824, 833 (9th Cir.1999). In Mezas de Jesus , we held that a nine-level increase in offense level for an uncharged kidnaping had an extremely disproportionate effect on the defendant’s sentence relative to the offense of which he was convicted, and therefore the district court erred in failing to apply a clear and convincing evidence standard. 217 F.3d at 642-44 . In Hopper , we held that a seven-level increase in offense level based on acquitted conduct required application of the clear and convincing standard. 177 F.3d at 833 . Neither case set forth any “bright line” for determining when due process requires the clear and convincing standard.

In United States v. Valensia, 222 F.3d 1173 (9th Cir.2000), cert. granted and judgment vacated on other grounds by Valensia v. United States, — U.S. -, 121 S.Ct. 1222 , 149 L.Ed.2d 133 (2001), we reviewed the Ninth Circuit cases which have considered the due process issue and identified the various factors that we consider in determining whether due process requires the application of a heightened evidentiary standard:

Id. at 1182 (citations omitted); see also id. (stating that we have “considered the disparity between the sentence that could have been imposed under the initial sentencing guideline range and the sentence actually imposed on a case-by-case basis, without relying on any single factor as controlling”). These factors do not require application of a higher standard of proof here.

The first and second factors are inapplicable; it is undisputed that Johansson was sentenced within the maximum range and that he pled guilty to the offense for which his sentence was enhanced.

Johansson contends that the third factor requires a higher standard of proof because the facts offered by the government in support of the enhancement created a new offense requiring separate punishment, namely, making false statements involving the conscious or reckless risk of serious bodily injury. He argues that because application of the enhancement eliminated the possibility of no imprisonment or a split sentence, the enhancement created a new offense requiring a separate punishment of imprisonment; in other words, it carried “with it imposition of a mandatory term of imprisonment where none existed in the absence of such application.”

This argument misapprehends the third factor. The Sentencing Guide lines do not as a general matter “create new offenses requiring separate punishment.” Restrepo, 946 F.2d at 657 . A separate offense is not created because an enhancement increases the applicable Guideline range beyond what it would have been without the enhancement; such an increase results from nearly every enhancement. Moreover, the Supreme Court has held that “consideration of information about the defendant’s character and conduct at sentencing does not result in ‘punishment’ for any offense other than the one of which the defendant was convicted.” Witte v. United States, 515 U.S. 389, 401 , 115 S.Ct. 2199 , 132 L.Ed.2d 351 (1995). Rather, the defendant is “punished only for the fact that the present offense was carried out in a manner that warrants increased punishment.” Id. at 402 , 115 S.Ct. 2199 .

The fourth factor — whether the enhancement is based on the extent of a conspiracy — is based on United States v. Harrison-Philpot, 978 F.2d 1520 (9th Cir.1992). In that case the defendant was convicted of seven counts of distributing a total of 67 grams of cocaine and one count of conspiracy to distribute cocaine. Id. at 1522 . The offense level for that quantity of cocaine, in light of the defendant’s criminal history category, resulted in a Guideline range of 41-51 months. The district court found that the defendant conspired to distribute between 15 and 49.9 kilograms of cocaine during the conspiracy and accordingly increased his offense level based on this amount resulting in a range of 292-365 months. The court sentenced the defendant at the high end of the range. We affirmed the district court’s application of the preponderance standard to the facts supporting the enhancement and held that due process does not require the application of the clear and convincing standard when “the extent of the conspiracy caused the tremendous increase in sentence.” Id.

Here, putting aside for the moment whether the § 2F1.1(b)(6)(A) enhancement caused a “tremendous increase” -in Johans-son’s sentence, the increase in sentence, like the increase in sentence in Harrison-Philpot , was caused by the extent of the conspiracy to which he pled guilty. Indeed, he stipulated that he “agreed to allow Atlas drivers to drive in excess of hours of duty prescribed by U.S. DOT and to conceal such violations by falsifying records required to be maintained by U.S. DOT, including records of duty status' (commonly referred to as ‘driver logs’ or ‘spread sheets’).” The issue at sentencing was the extent of that agreement and whether Johansson’s conduct, that is, his conspiracy, involved the conscious or reckless risk of serious bodily injury. The fact that the enhancement was based on a “specific offense characteristic” further demonstrates that it was based on the extent and nature of Johansson’s conspiracy. See U.S.S.G. § 2F1.1(b) (specific offense characteristics of offenses involving fraud or deceit).

The fifth factor is whether the increase in the number of offense levels is less than or equal to four. The parties characterize the enhancement as providing for a five-level increase. In the plea agreement they stipulated to a base offense level of six and to a two-level enhancement for more than minimal planning. Next, the government reserved the right to argue the § 2F1.1(b)(6)(A) enhancement for an offense that involved the conscious or reckless risk of serious bodily injury. Under that Guideline, the offense level is increased by two unless the resulting offense is less than 13. If it is less, the offense level is increased to 13. That “jump” in offense levels — from level eight to level 13 — is why the parties characterize the enhancement as providing for a five-level increase. The fact that the increase was more than four levels, however, means merely that Johansson’s due process claim does not fail on the ground that his offense level was increased by four or fewer levels. See Hopper, 177 F.3d at 833 .

In any event, Johansson’s offense level was, in fact, increased by four or fewer levels. The plea agreement also called for a four-level increase for being a leader, organizer, manager, and supervisor in a criminal activity involving five or more participants, see U.S.S.G. § 3B1.1(a), resulting in a stipulated offense level of 17. After a three-level reduction for acceptance of responsibility, Johansson’s total offense level was 14. Without the § 2F1.1(b)(6)(A) enhancement his offense level would have been 10 (base offense level of six; two levels for more than minimal planning; four levels for being a leader; less two levels for acceptance of responsibility). Thus, the enhancement ultimately increased his offense level by four. This four-level difference is the one most relevant to the due process inquiry; the ultimate increase in offense level caused by the enhancement. The fact that at some point in the calculation the offense level was increased by five levels does not tell us how disproportionate Jo-hansson’s actual sentence was to the offense to which he pled guilty.

Johansson also argues that the sixth factor supports application of the higher standard: “is the length of the enhanced sentence more than double the length of the sentence authorized by the initial sentencing guideline range in a case where the defendant would otherwise have received a relatively short sentence?” Valensia, 222 F.3d at 1182 . Johansson’s Guideline range would have been six to 12 months without the enhancement. With the enhancement he had a range of 15-21 months. Thus, the enhancement did not cause the sentencing range to double. Compare with Mezas de Jesus, 217 F.3d at 643 (21-27 months increased to 57-71 months); Hopper, 177 F.3d at 832 (24-30 months increased to 63-78 months). Moreover, the district court sentenced Johansson at the bottom of the range, to 15 months. Fifteen months is only three months more than 12 months — the maximum sentence authorized under the Guidelines without the enhancement — and it is only nine months more than the authorized minimum sentence; again, the length of the sentence ultimately imposed by the court was not twice the sentence authorized without the enhancement. Accordingly, the sixth factor does not warrant application of a higher standard of proof.

Johansson nonetheless urges us to consider “the dramatic increase in severity of punishment” implicated in increasing his offense level from 10 to 14. He emphasizes that at level 10 the district court would have had discretion to impose alternatives to imprisonment whereas at level 14 it had no discretion to do so.

We are unpersuaded. The critical question is whether Johansson’s due process rights were violated. Johansson pled guilty and stipulated to enhancements which rendered him eligible for a term of imprisonment of 12 months. A sentence of three months more than the sentence he agreed could be imposed is not extremely disproportionate to his offense of conviction.

Johansson also offers no compelling reason why the government should have a higher standard of proof for a defendant who would be eligible for a sentence that does not include imprisonment if a particular disputed enhancement is eliminated, than for a defendant whose offense level is initially subject to a mandatory term of imprisonment. Such a rule would place a greater burden on the government to prove enhancements for defendants who do not have any prior criminal conduct than on those who do. Due process does not require such a distinction.

The nature of the sentencing proceedings and the quality of the evidence presented by the government further compel rejection of Johansson’s constitutional argument. The higher standard of proof, after all, is meant to ensure that a defendant’s due process right to “an accurate application of the Guidelines” is met. Restrepo, 946 F.2d at 659 . The more process a defendant is afforded, including an evi-dentiary hearing with the right to cross-examination, the less likely a defendant’s right to “an accurate application of the Guidelines” may have been violated. Our due process analysis in Mezas de Jesus , for example, turned, in part, on the quality of the evidence upon which the district court based the enhancement. In discussing the unreliability of the hearsay statements linking the defendant to the uncharged kidnaping, we noted that “where we have found that the preponderance of the evidence standard sufficiently protected a defendant’s due process interests at sentencing, the evidence considered by the sentencing court was reliable and thoroughly 'tested.’ ” 217 F.3d at 644 . Thus, whether the preponderance of the evidence standard is constitutionally sufficient in any given case will depend, at least in part, upon the reliability of the evidence presented and the nature of the sentencing proceedings.

In this case the district court was extraordinarily careful in making sentencing findings. At the initial sentencing hearing it refused to apply the enhancement based only on the government’s written submission, and continued the hearing to give the parties an opportunity to further develop the evidence. It resumed the hearing two months later, and over the course of three days heard live testimony from five different witnesses, called by both parties. Jo-hansson submitted no fewer than three sentencing memoranda, and at the conclusion of the evidence each side had the opportunity, once again, to argue. This procedure is entirely consistent with the process envisioned by U.S.S.G. § 6A1.3.

In light of the above, the district court did not err in applying the preponderance of the evidence standard of proof to determine Johansson’s offense conduct. First, Johansson’s actual sentence was not extremely disproportionate to the offense of conviction. Second, unlike Hopper and Mezas de Jesus , his offense level was increased because of the nature and extent of the offense to which he pled guilty, rather than for acquitted or uncharged crimes. Third, the district court made its findings based upon three days of testimony, with direct and cross-examination, dozens of exhibits, several memoranda, and oral argument. In sum, the increase in Johansson’s offense level for committing a fraud that involved the conscious or reek- less risk of serious bodily injury was not “a tail which wags the dog of the substantive offense.” McMillan v. Pennsylvania, 477 U.S. 79, 88 , 106 S.Ct. 2411 , 91 L.Ed.2d 67 (1986).

Ill

Johansson next argues that even if the preponderance of the evidence standard applies, the evidence did not support application of the § 2F1.1(b)(6)(A) enhancement. We review a district court’s interpretation of the Sentencing Guidelines de novo, and review a district court’s factual findings in the sentencing phase for clear error. See United States v. Garcia, 135 F.3d 667, 669 (9th Cir.1998). We review a district court’s application of the Guidelines to the facts of a particular case for an abuse of discretion. See id.

We have never addressed the enhancement for a fraud offense involving the conscious or reckless risk of serious bodily injury. The enhancement grew out of the Major Fraud Act of 1988. The Act directed the Sentencing Commission to “provide for appropriate penalty enhancements, where conscious or reckless risk of serious personal injury has occurred.” Pub.L. 100-700, Chapter 47 sec. 2(b), Nov. 19, 1988, 102 Stat. 4632 . In response, the Commission adopted § 2F1.1(b)(6)(A). See United States v. McCord, 143 F.3d 1095, 1097 (8th Cir.1998). The first issue we must decide is precisely what the government must prove to support the enhancement, and, in particular, what is a “conscious risk” or “reckless risk” of serious bodily injury. The Commentary to section 2F1.1 does not provide any guidance.

The meaning of “reckless” risk of serious bodily injury was explored by the Eighth Circuit in McCord, supra. First, the court identified the normal meaning of reckless in the criminal context, namely, “that the defendant disregarded ‘a risk of harm which he is aware.’ ” 143 F.3d at 1097 (citing Fanner v. Brennan, 511 U.S. 825, 837 , 114 S.Ct. 1970 , 128 L.Ed.2d 811 (1994)). Next, the court noted that the Guideline for involuntary manslaughter adopts the criminal law definition of the term reckless:

Id. (quoting U.S.S.G. § 2A1.4, Commentary (n.l)). The court then concluded (without further elaboration) that the Commission intended the above definition of “reckless” to apply to “reckless” as used in § 2F1.1(b)(6)(A). Accordingly, it held that “the government must prove not only that the fraudulent conduct created a risk of serious bodily injury, but also that each defendant was in fact aware of and consciously or recklessly disregarded that risk.” Id. at 1098.

We apply the rules of statutory construction when interpreting the Sentencing Guidelines. See United States v. Robinson, 94 F.3d 1325, 1328 (9th Cir.1996). Our concern with the Eighth Circuit’s interpretation of “reckless” in the context of § 2F1.1(b)(6) (A) is that there is no meaningful distinction between an offense that involves the “conscious” risk of injury, and an offense that involves the “reckless” risk of injury, if under either prong the defendant must have been aware of the risk in the first place. The use of the disjunctive conjunction “or” between “conscious” and “reckless,” however, suggests that a conscious risk of injury is something different from and an alternative to a reckless risk of injury. See Hodg- son v. The Prophet Co., 472 F.2d 196, 206 (10th Cir.1973) (stating that the use of the word “or” between two phrases in a statute suggests that Congress intended that the two phrases be alternatives).

The Eighth Circuit appears to distinguish between a “conscious risk” and a “reckless risk” by holding that an offense involves the “conscious risk” when a defendant is aware of the risk but consciously disregards the risk, and an offense involves a “reckless risk” when the defendant is aware of the risk but recklessly disregards the risk. Such an interpretation, however, deviates from the plain language of the Guideline. The Guideline describes a “reckless risk,” not a reckless disregard of a known risk. We do not believe that a defendant can escape the application of the serious risk of injury enhancement by claiming that he was not aware that his conduct created a serious risk, that is, a defendant does not have to subjectively know that his conduct created the risk. See United States v. Vivit, 214 F.3d 908, 920 (7th Cir.), cert. denied, — U.S.-, 121 S.Ct. 388 , 148 L.Ed.2d 299 (2000) (holding that the district court properly applied the § 2Fl.l(b)(6)(A) enhancement and explaining that the issue is “whether the defendant’s fraudulent course of conduct created a risk that others would suffer serious bodily injury”); United States v. Hoffman, 9 F.3d 49, 50 (8th Cir.1993) (holding that under § 2F1.1(b)(6)(A) the government does not have to prove that the defendant intended serious harm, but only that he intended the conduct that created the risk of serious bodily injury). We therefore decline to adopt the Eighth Circuit’s interpretation of the conscious or reckless risk of serious bodily injury enhancement, and in particular, its definition of a “reckless risk.”

Nevertheless, we need not decide whether Johansson’s offense involved a reckless risk of injury because we have little difficulty concluding that a preponderance of the evidence supports the district court’s finding that his offense involved the conscious risk of serious bodily injury. It is apparent from the nature of Johansson’s offense itself — creating false logbooks to conceal hours-of-driving violations — that the offense involved the risk of serious bodily injury. The regulations governing the log books and the hours-of-driving requirements are entitled “Federal Motor Carrier Safety Regulations — Hours of Driving Drivers.” C.F.R. Title 49, Part 395 (emphasis added). The hours-of-driving limitations are plainly designed to limit driver fatigue and therefore reduce motor carrier accidents. Violations of those regulations therefore create a “risk” .of truck accidents and serious bodily injury. Moreover, by concealing the hours-of-driving violations by creating false log books, Johansson magnified the risk created by the violations by ensuring that they would continue undetected. Thus, Johansson’s fraudulent scheme appears to involve an obvious risk of serious bodily injury. See United States v. Laughlin, 26 F.3d 1523, 1530-31 (10th Cir.1994) (holding that fraudulent medical billing scheme that resulted in unnecessary surgical procedure involved the conscious or reckless risk of serious bodily injury); United States v. Hoffman, 9 F.3d 49, 50 (8th Cir.1993) (holding that fraudulent insurance scheme of staging automobile accidents involved the conscious or reckless risk of serious bodily injury even though defendant arranged only “slow speed” accidents).

In any event, the government offered more than the regulations themselves as evidence of the risk of injury created by Johansson’s fraudulent scheme. Its evidence included a transcript of the September 1999 testimony of Kenneth M. Mead, Inspector General of the DOT, before a congressional subcommittee on transportation, as well as materials from a group known as “Parents Against Tired Drivers.” The testimony and materials document the serious risk of truck accidents caused by violations of the hours-of-driving requirements. Amy Hope, a DOT Inspector, also testified that a carrier’s falsification of log books significantly affects a carrier’s safety rating, especially with carriers of hazardous materials such as Atlas. A carrier’s failure to correct an unsatisfactory safety rating could lead the DOT to stop the carrier’s driving. Thus, the falsification of log books impedes the DOT’s ability to reduce the risk of serious truck accidents by shutting down motor carriers that repeatedly violate safety regulations.

The key to the district court’s decision, however, was its finding that the hours-of-driving violations and false statements to conceal the violations were a regular part of Johansson’s business. This finding was not clearly erroneous. The district court found that Surprenant’s testimony conveyed the impression that the violations happened more than once in awhile. The government also proffered the statements of Virginia Spence, who served as Atlas’ Safety Manager from 1991 to 1996, and who stated that the violations occurred on a daily basis and were not sporadic. The statements of various Atlas employees who admitted to hours-of-driving violations also supported the district court’s finding. James Bolla, an Atlas Safety Manager from approximately 1993 until 1995, testified that ten percent of the Atlas drivers he audited were engaged in hours-of-driving violations and falsification of log books.

The district court’s finding that Johans-son was aware of the violations and therefore aware of the safety risk was also not clearly erroneous. The record includes reliable evidence that Johansson personally directed his managers to engage in the conduct, that he personally authorized cash payments to drivers for their hours worked in excess of the hours-of-driving regulations, and that he directed that the information as to the hours-of-driving violations not be reported to Atlas’ Safety Department. Moreover, the DOT advised Johansson of his companies’ hours-of-driving and daily log violations in 1994 and again in April 1996. The violations nonetheless continued until a search warrant was executed at Atlas in July 1996.

The evidence also supports the district court’s finding that Johansson ignored the safety risk created by the persistent hours-of-driving violations. He did not stop the practice even after twice being advised by the DOT that the violations were occurring; he directed managers to engage in the practice; and he instructed his subordinates to conceal the accurate hours-of-driving information from the Atlas Safety Department.

Johansson nonetheless argues that the evidence is insufficient because Atlas had a better than average safety record and there is no evidence that a driver was ever fatigued on any particular occasion. The Guideline, however, requires only that the offense involved the “risk” of serious bodily injury. See United States v. Turner, 102 F.3d 1350, 1358-59 (4th Cir.1996) (holding that fact that no serious injuries were attributable to the defendant’s failure to provide miners with required safety training did not preclude finding that the failure to provide the training created a risk of serious bodily injury). Johansson’s business practice of permitting hours-of-driving violations by operators of commercial trucks carrying hazardous materials created a risk that a driver would become fatigued and cause an accident resulting in serious bodily injury to the driver or others on the road.

Johansson also insists that most of his operators drive intrastate routes and therefore their hours-of-driving violations were not within the jurisdiction of the DOT and thus may not be considered in determining whether Johansson’s conduct involved the conscious or reckless risk of serious bodily injury. Johansson, however, pled guilty to conspiring to violate the hours-of-driving regulations and making false statements in matters within the jurisdiction of the DOT, and there is no dispute that some of his operators drive interstate routes. In light of the extent of the conspiracy, the evidence is sufficient to support a finding that the interstate hours-of-driving violations involved the conscious or reckless risk of serious bodily injury.

Finally, Johansson offered evidence that hours-of-driving violations and false record keeping are widespread in the trucking industry. This evidence suggests that we should be concerned about the safety of driving on the nation’s highways; it does not suggest that Johansson’s conduct did not involve the conscious or reckless risk of serious bodily injury.

Johansson lastly contends that Apprendi v. New Jersey, 530 U.S. 466 , 120 S.Ct. 2348 , 147 L.Ed.2d 435 (2000), requires reversal of his sentence because the district court’s increase in his offense level pursuant to §’ 2F1.1(b)(6)(A) increased his sentence “beyond the de facto maximum established in the plea agreement.” In other words, because Johansson did not stipulate to the facts upon which the increase in his sentence was based, those facts had to be found by a jury beyond a reasonable doubt.

Because Johansson did not raise the Apprendi issue before the district court (Apprendi had not been decided), the appellate court reviews the issue for “plain error.” Fed.R.Crim.P. 52(b); United States v. Pacheco-Zepeda, 234 F.3d 411, 413 (9th Cir.2000), as amended on rehearing, 234 F.3d 411 (9th Cir.2000). Under the plain error doctrine, Johansson must show that (1) an error was committed, (2) the error was “plain,” and (3) the error affected Johansson’s substantial rights. See United States v. Nordby, 225 F.3d 1053, 1059 (9th Cir.2000). If these conditions are met, the court may exercise its discretion to review the error only if it “ ‘seriously affect[s] the fairness, integrity, or public reputation of judicial proceedings.’ ” Id. (citation omitted).

In Apprendi the Supreme Court held that “any fact (other than a prior conviction) that increases the maximum penalty for a crime must be charged in an indictment, submitted to a jury, and proved beyond a reasonable doubt.” 120 S.Ct. at 2362-63 . The increase in offense level at issue, however, did not increase the maximum penalty for the crime to which Johansson pled guilty; namely, conspiracy to make false statements in violation of 18 U.S.C. § 371 and § 1001. The statutory maximum penalty for conspiracy is five years and there is no maximum for section 1001. Johansson was sentenced to 15 months; he does not have an Apprendi claim. See United States v. Hernandez-Guardado, 228 F.3d 1017, 1027 (9th Cir.2000) (holding that Apprendi did not apply to two-level enhancement that resulted in a sentence that did not exceed the ten-year statutory maximum).

AFFIRMED.

. The enhancement is presently codified at U.S.S.G. § 2F1.1 (b)(7)(A). . The Information alleged that the hours-of-driving and daily log requirements were part of the DOT regulations entitled "Transportation of Hazardous Materials; Driving and Parking Rules.” Those regulations, however, merely provide that operators of motor vehicles transporting hazardous materials must comply with federal motor carrier safety regulations. 49 C.F.R. § 397.2 . The DOT has not adopted special hours-of-driving and daily log regulations for drivers of hazardous materials; the regulations at issue apply to all "motor carriers," including those that are not transporting hazardous materials. See United States v. McCord, Inc., 143 F.3d 1095, 1096 (8th Cir.1998) (applying hours-of-driving and daily log regulations to motor carrier). . The Guidelines recognize the importance of the reliability of the evidence and the process afforded during a sentencing proceeding: When any factor important to the sentencing determination is reasonably in dispute, the parties shall be given an adequate opportunity to present information to tire court regarding that factor. In resolving any dispute concerning a factor important to a sentencing determination, the court ■may consider relevant information without regard to its admissibility under the rules of evidence applicable at trial, provided that the information has sufficient indicia of reliability to support its probable accuracy. § 6A1.3(a); see also id. (Commentary) (“An evidentiary hearing may sometimes be the only reliable way to resolve disputed issues”); id. (Commentary) ("The sentencing court must determine the appropriate procedure in light of the nature of the dispute, its relevance to the sentencing determination, and applicable case law”). . Legislative history also supports a distinction between a “conscious risk” and a “reckless risk.” The Major Fraud Act of 1988 added a new criminal statute for major fraud against the United States. 18 U.S.C. § 1031 . Section 2(a) of the Act included enhanced penalties for a section 1031 offense that “involves a conscious or reckless risk of serious personal injury.” (Section 2(b) of the Act directed the Sentencing Commission to adopt a guideline for fraud involving a conscious or reckless risk of serious bodily injury.) The Senate Report explains that the “term ‘conscious’ means the defendant knew of the risk,” and that “reckless” should be "interpreted consistently with the generally understood requirements for a finding of recklessness or criminal negligence.” Senate Report No. 100-503, Sep. 12, 1988. This legislative history is evidence that Congress intended that a "conscious risk,” that is, a risk of which the defendant was aware, is something different from a reckless risk.