O’SCANNLAIN, Circuit Judge:
We must decide whether the sentencing provision of the statute criminalizing forged endorsements of United States Treasury checks, 18 U.S.C. § 510, is sufficiently ambiguous to warrant application of the rule of lenity.
I
Tami LeCoe is the second wife of Joseph E. LeCoe, a naval recruiter in Boise, Idaho. She married Joseph on April 1, 1983, two months after his divorce from his first wife, Pamela J. LeCoe.
The decree finalizing the divorce of Joseph and Pamela required Joseph to make eight monthly payments of $400 each to Pamela for her share of his military retirement. In addition, Joseph was ordered to make child support payments of $300 per month for the duration of the minority of the couple’s two children. Joseph arranged to have these payments taken directly from his retirement pay; accordingly, the checks to Pamela were issued by the United States Treasury. For reasons not relevant to these proceedings, the checks to Pamela for the period February 1983 to September 1984 were sent to Joseph, who was in turn responsible for forwarding them to Pamela. Beginning in October 1984, Joseph arranged to have the checks mailed directly to Pamela.
Pamela eventually complained to authorities that she had received only five of the twenty-one checks due her from February 1983 to September 1984. In addition, she complained that she had not received her share of the couple’s 1982 income tax refund. An investigation ensued, in which it was discovered that Tami had forged Pamela’s name to several of the checks and [400]*400deposited them into the joint account of Joseph and Tami. Tami had also forged Pamela’s name onto a $600 cashier check, which represented Pamela’s share of the 1982 income tax refund.
A federal grand jury handed down a twelve-count indictment against Tami on November 20, 1987. She was charged with five violations of 18 U.S.C. § 495,1 and seven violations of 18 U.S.C. § 510.2 Trial before a jury commenced in January 1990, resulting in Tami’s conviction on four of the section 510 counts. Tami was found guilty of forging Treasury checks issued to Pamela in December 1983 and February, April, and May 1984. Each of the first three checks was for $400; the fourth was for $300. She was acquitted on all remaining counts.
Sentencing was scheduled for March 27, 1990. The court first considered Tami’s motion to classify the convicted offenses as misdemeanors rather than felonies.3 The court denied the motion, reiterating that the four convictions would be treated as felony convictions. The district court then suspended sentence for all four counts in lieu of a three-year term of probation. Tami was also ordered to pay restitution of $1,500.
This appeal followed.4
II
Prior to November 1983, forged endorsements of Treasury checks were prosecuted under 18 U.S.C. § 495, which penalizes anyone who “falsely makes, alters, forges, or counterfeits any deed, power of attorney, order, certificate, receipt, contract, or other writing” with the intent of defrauding the United States. Although section 495 does not specifically outlaw forged endorsements of Treasury checks, such forgery has been deemed illegal under the “other writing” provision of the section. See Prussian v. United States, 282 U.S. 675, 679-80, 51 S.Ct. 223, 225, 75 L.Ed. 610 (1931) (clause “other writing” was intended to “extend[] the penal provisions of the statute to all writings of every class if forged for the purpose of obtaining money from an officer of the United States”). Violation of section 495 is a felony, punishable by imprisonment for up to ten years, a fine not to exceed $1,000, or both.5
Section 495, however, did not adequately encompass the scope of culpable activity surrounding the wrongful cashing of Treasury checks. Senator DeConcini, when introducing remedial legislation, explained:
Presently, it is possible for a thief to steal a Treasury check endorsed by a payee, endorse his own name and obtain the proceeds without violating section 495. It is also possible for a thief to steal one or more Government checks or bonds and sell or exchange them to a middle man and not violate section 495. These kind of situations require the prosecutor to “hunt around” for a statute which often does not fit the crime. As a result, these cases are often either not prosecuted or the charges are dismissed.
129 Cong.Rec. S9342 (Statement of Sen. DeConcini). Accordingly, Senator DeConci-ni sponsored a bill to rectify these shortfalls. The bill, part of what soon became Public Law 98-151, is now codified at 18 U.S.C. § 510. Section 510, in short, crimi[401]*401nalizes forged endorsements of Treasury checks, bonds, or United States securities, including the fraudulent conduct not captured by section 495. As a general rule, a person convicted under section 510 may be imprisoned for up to ten years, fined up to $10,000, or both.
Section 510 also modified the law regarding forged endorsements of Treasury checks in another important respect. Specifically, subsection (c) provides that violations of the section that do not exceed $500 should be prosecuted as misdemeanors. The subsection provides in full that:
If the face value of the Treasury check or bond or security of the United States or the aggregate face value, if more than one Treasury check or bond or security of the United States, does not exceed $500, in any of the above-mentioned offenses, the penalty shall be a fine of not more than $1,000 or imprisonment for not more than one year, or both.
While section 510 resolved several problems, it created some of its own. First, section 510 is silent as to whether it effected a partial repeal of section 495; resolution of this question was left to the courts. To date, the circuits that have considered this issue have unanimously concluded that section 510 was not intended as a partial repeal. See United States v. Edmonson, 792 F.2d 1492, 1497-98 (9th Cir.1986), cert. denied, 479 U.S. 1037, 107 S.Ct. 892, 93 L.Ed.2d 844 (1987); United States v. Oliver, 908 F.2d 260, 264 (8th Cir.1990); United States v. Barrett, 837 F.2d 933, 934-35 (10th Cir.1988); United States v. Cavada, 821 F.2d 1046, 1047 (5th Cir.), cert. denied, 484 U.S. 932, 108 S.Ct. 304, 98 L.Ed.2d 263 (1987); Edwards v. United States, 814 F.2d 486, 489-90 (7th Cir.1987). Thus, a forger of Treasury check endorsements may be prosecuted under either statute; a defendant who may qualify for misdemeanor prosecution under section 510 may therefore be prosecuted for a felony under section 495. See Edmonson, 792 F.2d at 1498; Oliver, 908 F.2d at 264; Cavada, 821 F.2d at 1047-48.
In addition to the partial repeal question, some litigants have questioned the clarity of section 510 s sentencing provisions. Some, such as the appellant here, maintain that section 510 does not clearly indicate what checks may be aggregated to determine whether one qualifies for felony or misdemeanor prosecution under the section. One school of thought is that the value of all checks underlying an indictment may be aggregated, even if the charges in the indictment are unrelated and are separate offenses. Under this scheme, if the sum of any or all checks in any or all counts exceeds $500, then each individual section 510 violation in the indictment becomes a felony.
The contrasting viewpoint suggests that only checks underlying each individual count may be totalled. Thus, individual offenses that would otherwise be misdemeanors are not transmogrified into felonies simply because the sum total of all checks for all charged offenses exceeds $500.
To date, the question of subsection (c)’s clarity has resulted in only one published opinion. See United States v. Taylor, 869 F.2d 812 (5th Cir.), cert. denied, — U.S. —, 110 S.Ct. 171, 107 L.Ed.2d 128 (1989). In a split decision, the Fifth Circuit concluded that both “common sense” and “plain meaning” of the statute required reading section 510 as “permitting] the court to aggregate the value of all checks that form the basis of a violation no matter whether the checks form the basis of one or of many chargeable offenses.” Id. at 815. The court reasoned that since subsection (c) was itself a lenity provision, there was essentially no need for further leniency. See id. at 814. Moreover, the court noted that a contrary reading of section 510 would enable “crafty” or “habitual” criminals to commit repeated violations in small amounts without fear of felony prosecution. Id. at 815.
Judge Thornberry dissented. He believed that the statute was sufficiently ambiguous to require invocation of the rule of lenity. Id. at 816 (Thornberry, J., dissenting). Moreover, he noted that, as a matter of policy, it was unreasonable to believe that Congress intended to permit felony [402]*402prosecution for minor offenses remote in both time and place. Id. at 817. Thus, Judge Thornberry adopted the view that only check amounts for each chargeable offense could be aggregated.
With this background, we turn to the merits of this appeal. We review questions of statutory construction and interpretation de novo. See United States v. Valencia-Roldan, 893 F.2d 1080, 1082 (9th Cir.), cert. denied, — U.S. —, 110 S.Ct. 2181, 109 L.Ed.2d 509 (1990).
Ill
A
LeCoe contends that subsection (c) is ambiguous and, accordingly, must be construed in her favor under the rule of lenity. The rule of lenity provides that “ambiguity concerning the ambit of criminal statutes should be resolved in the favor of lenity.” Rewis v. United States, 401 U.S. 808, 812, 91 S.Ct. 1056, 1059, 28 L.Ed.2d 493 (1971). The Supreme Court has identified two policies underlying this rule. First, concerns of fairness suggest that “warning should be given to the world in language that the common world will understand, of what the law intends to do if a certain line is passed. To make the warning fair, so far as possible the line should be clear.” United States v. Bass, 404 U.S. 336, 348, 92 S.Ct. 515, 522, 30 L.Ed.2d 488 (1971) (quoting McBoyle v. United States, 283 U.S. 25, 27, 51 S.Ct. 340, 341, 75 L.Ed. 816 (1931) (Holmes, J.)). Second, “because of the seriousness of criminal penalties, and because criminal punishment usually represents the moral condemnation of the community, legislatures and not courts should define criminal activity.” Id. at 348, 92 S.Ct. at 523.
The lenity doctrine has been extended beyond interpretations of the substantive ambit of criminal prohibitions; the doctrine also encompasses the penalties imposed by criminal statutes. See Hughey v. United States, — U.S. —, 110 S.Ct. 1979, 1985, 109 L.Ed.2d 408 (1990) (applying rule of lenity to restitution provision of the Victim and Witness Protection Act, 18 U.S.C. §§ 3579-3580). “This policy of lenity means that the Court will not interpret a federal criminal statute so as to increase the penalty that it places on an individual when such an interpretation can be based on no more than a guess as to what Congress intended.” Bifulco v. United States, 447 U.S. 381, 387, 100 S.Ct. 2247, 2252, 65 L.Ed.2d 205 (1980) (quoting Ladner v. United States, 358 U.S. 169, 178, 79 S.Ct. 209, 214, 3 L.Ed.2d 199 (1958)).
Despite its lofty ideals, the rule of lenity is not an automatic addendum that accompanies every criminal statute that Congress may choose to enact; rather, it is simply a canon of statutory construction. See United States v. Rodgers, 466 U.S. 475, 484, 104 S.Ct. 1942, 1948, 80 L.Ed.2d 492 (1984). The rule plays no role in statutory interpretation unless the statute is truly ambiguous. See Moskal v. United States, — U.S. —, 111 S.Ct. 461, 465, 112 L.Ed.2d 449 (1990).
Thus, we must first determine whether section 510 is ambiguous. A statute is not ambiguous simply because it is possible to construe a statute narrowly. See id. Nor does a division of judicial authority necessarily trigger the rule. See id. Rather, the rule of lenity is reserved “for those situations in which a reasonable doubt persists about a statute’s intended scope even after resort to the language and structure, legislative history, and motivating policies of the statute.” Id. (emphasis in original; quotation omitted). Accordingly, we consider each of these factors in turn.
B
In determining the scope of a statute, a court must look first to the statute’s language and structure. See id. 111 S.Ct. at 465; see also United States v. 594,464 Pounds of Salmon, 871 F.2d 824, 825 (9th Cir.1989). If the statute’s language is unambiguous, its plain language controls except in rare and exceptional circumstances. See 594,464 Pounds of Salmon, 871 F.2d at 826; see also Howe v. Smith, 452 U.S. [403]*403473, 483, 101 S.Ct. 2468, 2475, 69 L.Ed.2d 171 (1981). The words used in the statute are given their ordinary meaning. Moskal, 111 S.Ct. at 465.
The contentious portion of subsection (c) reads: “If the face value of the Treasury check or bond or security of the United States or the aggregate face value, if more than one Treasury cheek or bond or security of the United States, does not exceed $500_” (emphasis added). The Fifth Circuit concluded that the “plain meaning” of this provision was clear; a court could “aggregate the value of all checks that form the basis of a violation no matter whether the checks form the basis of one or of many chargeable offenses.” Taylor, 869 F.2d at 815 (emphasis added). The Taylor Court’s reasoning apparently stemmed from the assumption that Congress would have qualified the aggregation clause if it had intended a court to aggregate less than every check value before it in the particular case, regardless of the actual number of separate alleged counts. Simply put, the “whole” should be presumed unless a subset is clearly specified.
The Fifth Circuit also relied upon the structure of the statute in reaching its conclusion. Subsection (a) of section 510 both describes the conduct outlawed by the subsection—forged endorsement of, inter alia, Treasury checks—and, in the same sentence, establishes the general penalty for committing such outlawed conduct, e.g., imprisonment for up to ten years, a fine up to $10,000, or both. Subsection (c), in contrast, does not purport to describe any proscribed conduct. Rather, it sets forth an exception to the penalty provisions of the prior two subsections: if the face value of the check or checks involved does not exceed $500, then the maximum sentence permitted is imprisonment for up to one year or a fine not to exceed $1,000. Subsection (c), then, is an exception to the general penalty provisions, leading the Fifth Circuit to conclude that “[s]ubsection (c) is itself a lenity provision, designed to spare less than major offenders the rigors of the general punishment provision.” Taylor, 869 F.2d at 814. The court implicitly concluded, then, that further lenient reading of the subsection would unnecessarily and, presumably, wrongfully expand congressional generosity beyond its intended scope.
The Fifth Circuit’s rationale is superficially appealing. However, the court’s conclusions are grounded in two unspoken assumptions: first, that Congress would have expressly so provided if it intended less than “all” checks in an indictment to be aggregated and, second, that Congress’ apparent motivation towards leniency precludes any further review of its motivations in construing the subsection. It would be premature for us to adopt the Taylor majority’s reasoning without examining the validity of these underlying assumptions.
We begin with the Taylor Court’s first assumption.
It is axiomatic that “[ejach count in an indictment is regarded as if it was a separate indictment.” Dunn v. United States, 284 U.S. 390, 393, 52 S.Ct. 189, 190, 76 L.Ed. 356 (1932) (Holmes, J.); see also Selvester v. United States, 170 U.S. 262, 267, 18 S.Ct. 580, 581, 42 L.Ed. 1029 (1898). Accordingly, it has been held that each count in an indictment must stand on its own, and cannot base its validity on the allegations of any other count not specifically incorporated. See, e.g., United States v. Olatunji, 872 F.2d 1161, 1166 (3d Cir.1989); United States v. Winter, 663 F.2d 1120, 1138 (1st Cir.1981), cert. denied, 460 U.S. 1011, 103 S.Ct. 1250, 75 L.Ed.2d 479 (1983); United States v. Fulcher, 626 F.2d 985, 988 (D.C.Cir.), cert. denied, 449 U.S. 839, 101 S.Ct. 116, 66 L.Ed.2d 46 (1980).
Congress is, of course, presumed to know existing law pertinent to any new legislation it enacts. Native Village of Venetie v. Alaska, 918 F.2d 797, 803 (9th Cir.1990). Thus, we can presume that Congress knew of the sanctity bestowed upon each individual count within an indictment by the Supreme Court, and that as a result of this sanctity, each individual count within an indictment must stand on its own. Of [404]*404course, the statute at issue here concerns aggregation for purposes of sentencing rather than for establishing an essential element of the substantive offense and thus the analysis in Dunn is not squarely on point. However, the point of this analysis is simple: given the sanctity which the courts, including the Supreme Court, have accorded each individual count, it is as fair to conclude that Congress would assume a single count as the proper unit for aggregation as it is to assume that such aggregation would encompass all counts in an indictment.6
Accordingly, we must conclude that the language of subsection (c) is ambiguous.
To test the Taylor Court’s implicit assumption as to Congress’ policy underlying section 510(c), and to fulfill Moskal's multi-facet test for establishing statutory ambiguity, we next examine section 510’s legislative history.
Section 510’s legislative history is sparse. The bill that became section 510, Senate Bill 1558, was apparently an eleventh hour amendment to a much larger piece of legislation. Accordingly, much of section 510's history is the statement of the bill’s sponsor, Senator DeConeini. His statement is, however, illuminating.
Senator DeConeini identified six specific benefits associated with such a misdemean- or provision:
First. Result in more forgers being prosecuted by U.S. attorneys;
Second. Eliminate the need for creating “legal fictions” which result from prosecuting under misdemeanor laws that may not directly address the forgery offense;
Third. Increase plea negotiation flexibility and guilty pleas;
Fourth. Increase use of the U.S. [magistrate judges] in check forgery cases and reduce the number of felony trials, thereby reducing the burden on the Federal district courts;
Fifth. Provide more accurate conviction records which may be useful in future prosecutions of repeat offenders; and
Sixth. Provide a realistic penalty structure.
129 Cong.Rec. S9342 (Statement of Sen. DeConeini). None of these policies directly support a harsh reading of subsection (c). Indeed, most bespeak a broad application of the misdemeanor provision. Three of these benefits are, in particular, advanced by a broad reading of the statute; a broad reading of subsection (c) would presumably result in (1) more plea negotiation flexibility and, by extension, more guilty pleas, (2) increased use of federal magistrate judges in check forgery cases, and (3) a higher possibility for realistic penalties than would the strict reading of the subsection advanced by the government.
Section 510’s legislative history is, at best, inconclusive.
Lurking beneath our analysis is yet another canon of statutory construction: “[statutes are to be construed together to effectuate, to the greatest extent possible, the legislative policies of both.” Strobl v. New York Mercantile Exchange, 768 F.2d 22, 30 (2d Cir.), cert. denied, 474 U.S. 1006, 106 S.Ct. 527, 88 L.Ed.2d 459 (1985) (emphasis added). Here, felony convictions for forged endorsements of Treasury checks are generally possible under two separate [405]*405statutes, 18 U.S.C. § 495 and 18 U.S.C. § 510(a). As previously noted, a person may be convicted under the felony provision of section 495 even if he or she could only be tried for a misdemeanor under section 510. Thus, the Fifth Circuit’s fear of the crafty thief circumventing section 510 are largely illusory. Only in a rare instance will such a crafty thief be able to avoid felony prosecution under both sections 495 and 510. Thus, under a broad reading of subsection (c), prosecutors and courts are given maximum flexibility in structuring charges and sentences, all without hampering the prosecutor’s ability to bring felony charges in appropriate cases. The effectiveness of both statutes is maximized while the overlap between the two statutes is minimized.
C
In sum, the plain language of subsection (c) does not clearly specify what may, or may not, be aggregated for the purpose of determining eligibility for its misdemeanor provision. While it is possible to read the statute narrowly, the legislative history and motivating policies would suggest that the misdemeanor provision should be read broadly. Likewise, a broad reading is warranted when section 510 is viewed in conjunction with section 495, which permits felony convictions even in cases where only a misdemeanor could be charged under section 510. Thus, while none of these factors is dispositive, there can be no doubt that subsection (c) is, at best, ambiguous and the legislative history inconclusive. Accordingly, we are compelled to invoke the rule of lenity. Aided by this canon of statutory construction, we conclude that section 510(c) authorizes aggregation only of those checks that comprise a single charged offense.7
IV
LeCoe was convicted of four separate and distinct violations of section 510. Since none of the Treasury checks exceeded $500 in any of the four counts, either individually or aggregated per count, Le-Coe could properly only have been sentenced under the misdemeanor provision of section 510. Accordingly, the district court erred when it imposed sentence upon Le-Coe as if she had been convicted of four felonies.
REVERSED and REMANDED for re-sentencing.