Audit Clause Meets Attorney-Client Privilege
 Frequently, government auditors and investigators demand that contractors provide documents that attorneys consider to be subject to privilege, and thus protected. The situation creates conflict between what the auditors assert they need for their audit and what the attorneys assert they do not have to provide to anybody. Indeed, the attorneys assert that there are adverse consequences from providing material protected by the attorney-client privilege.
The matter often comes up in audits of consultants’ expenses, where invoices from external attorneys are requested so that they can be reviewed for allowability. The matter can arise in requests for internal audits and/or internal investigations conducted at the behest of corporate counsel. The matter almost always bubbles up in audits of contractor disclosures made pursuant to the 52.203-13 contract clause. The tension between evidentiary support and legal privilege comes up in other contexts as well.
And it’s almost always a problem.
The problem is that providing auditors or investigators with attorney-privileged material may act to waive that privilege with respect to downstream litigation. We wrote about that problem right here. We discussed a recent case where privilege was deemed to have been waived when an internal investigation was provided to the GSA Inspector General, and we opined that “Such reviews should only be provided to outsiders (including government personnel such as DCAA auditors or Agency Inspectors General) under very carefully considered circumstances, lest the door be opened for qui tam relators and opposing counsel to obtain a bounty of documents that would have otherwise been protected.”
More recently, a D.C. district court told KBR that its 89 internal fraud investigation reports (prepared under privilege and related to alleged violations of KBR’s ethics policy) were not protected by the attorney-client privilege. According to an analysis of the court’s decision, prepared by the attorneys at GreenbergTaurig—
… the court reasoned that because these investigations are required by federal procurement regulations, the company would have had to conduct them anyway, regardless of whether they had been advised to do so by an attorney. The court relied on the ‘mandatory’ nature of the investigations in reaching its conclusion that the documents were neither privileged nor protected by the work product doctrine.
KBR quickly appealed via a writ of mandamus, and the D.C. Circuit Court of Appeals reversed. In a unanimous decision, the Appellate Court found that the doctrine of attorney-client privilege protected KBR’s internal investigations. That was a fortunate outcome, both for KBR and for other government contractors.
The conflict arose again when two Department of Energy contractors at the Hanford Waste Treatment & Immobilization Plant (“WTP”) refused to provide the DOE Inspector General with more than 4,300 requested documents. Both the prime contractor (Bechtel National, Inc.) and its subcontractor (URS Energy and Construction, Inc.) allegedly refused to cooperate with the DOE IG’s investigation into the firing of a whistle-blower (Ms. Donna Busche). Allegedly, Ms. Busche was fired by URS “after raising safety concerns” regarding activities at the $12 Billion project to treat radioactive waste at the site, which played a key role in the World War II “Manhattan Project” and the “Cold War” arms race that followed.
The DOE IG spent six months investigating whether or not Ms. Busche was terminated in retaliation for her “blowing the whistle” on unsafe practices. According to the DOE IG, its investigators could not reach any conclusions “because of a material scope limitation.” In other words, because the DOE contractors refused to provide the requested documents, the auditors had nothing to audit. According to the DOE IG—
On the advice of outside counsel, both contractors took the position that the documents in question were subject to either attorney-client or attorney work product privilege. Also, URS made a unilateral determination that certain documents were not relevant to our examination. Specifically, Bechtel withheld 235 documents and URS withheld 4,305 documents. Of the 4,305 withheld documents, URS' attorney eventually agreed to provide access to a portion of the 2,754 documents that URS had concluded were non-responsive but which were not subject to the asserted attorney-client privilege. Attorneys representing both Bechtel and URS stated that the assertion of privilege was necessary given the likelihood of litigation regarding the Busche matter. Their basic concern was that releasing the documents to the Office of Inspector General would constitute a waiver of privilege in future proceedings.
The DOE IG told Bechtel and URS that their contracts with DOE contained clauses that required all documents “acquired or generated under the contract, including those for which attorney-client and attorney work product privilege were asserted.” However, the contractors’ attorneys didn’t agree with that interpretation of the clauses’ requirements. The DOE IG reported that “It was the position of counsel for both Bechtel and URS that these clauses were too broad and that they were unenforceable, specifically in situations where litigation was either in process or was likely.”
And the contractors’ attorneys stuck to that position, despite pressure applied to the companies.
That position did not sit well with Senator Claire McCaskill, (D-MO). You may remember Senator McCaskill from her grilling of then-DCAA Director April Stephenson. Ms. McCaskill reportedly “slammed” the two contractors, writing (in a letter to the Secretary of Energy) that she would like him to tell her how they would be held accountable “for their noncompliance, including withholding of fees and recovery of costs incurred” by the DOE IG.
Bechtel had its own take on the situation. It wrote—
Bechtel went above and beyond in cooperating with the OIG's investigation--providing requested documents for review and people to interview, in accordance with the protocol agreed to with the IG. Furthermore, we offered to work with the OIG to provide access to documents that are protected under the law, in a way that preserves those protections, but the OIG declined our offer.Â
And there the matter sits (so far as we know).
Meanwhile, news reports indicate that KBR is back in court, fighting a demand by a qui tam relator that it turn over documents provided to the government. The relator’s attorneys are arguing that KBR waived attorney-client privilege when it decided to provide the documents.
And so it goes …
Solving Executive Compensation Concerns with Blended Rates
The allowability of contractor executive compensation is a complex, tricky, thing—made more tricky by recent statutory and regulatory changes. We have written about some of those recent changes before.
At this point, the average contractor must handle three separate rules that each establish separate limits on allowable compensation. Some contracts are subject to the “old” executive compensation ceiling of $952,308, as applied to the Top 5 most highly compensated individuals in each segment. Other contracts are subject to the “old” ceiling as applied to all contractor employees (not just the Top 5). Still other (newer) contracts are subject to a lower compensation ceiling of $487,000, as applied to all contractor employees. The ceiling on allowable compensation depends on when the contract was issued and its effective date, because it is the FAR Part 31 cost principle language in effect on that effective date that establishes the applicable ceiling.
Based on the foregoing, it would seem logical that each contractor must establish three different sets of billing rates, based on those three different sets of allowability rules, and then apply the appropriate set of billing rates to the appropriate contract, based on establishing each individual contract’s effective date.
That seems to be a problematic approach, right? Think about it:
Three compensation allowability calculations.
Three sets of provisional billing rates.
Three sets of final billing rates.
And then you have to know which set applies to which contract. The forward pricing part might seem to be relatively easy to handle, because all new contracts will be subject to the most restrictive (lowest) compensation ceiling. Even so, the interim (provisional) billing rates for 2014 and beyond will have to account for the fact that there are several groups of contracts, each with its own allowability criteria. The same is true for the final (actual) billing rates for 2014 and beyond.
Think that’s tough to implement? Maybe so. In fact, we’ll assert that it is so. But that’s what seems to be called for by the regulations.
The whole situation seems more than a little risky. You had better not make any mistakes. At a minimum, DCAA might allege that you billed expressly unallowable compensation costs, or that you included expressly unallowable compensation costs in your proposal to establish final billing rates. There are penalties and interest payments associated with expressly unallowable costs.
If you prepare your forward pricing rates (or calculate estimated indirect rates in your cost proposals) and you don’t exclude the right amount of executive compensation, then you might be accused of submitting certified cost or pricing data that was not accurate – which would be a violation of what used to be called “TINA”.
It’s possible (though perhaps unlikely) that somebody might allege some kind of violation of the False Claims Act – i.e., that the contractor intentionally or negligently overbilled the government). That would be a bad thing, very bad.
This situation cannot be what the lawmakers intended when they lowered the allowable executive compensation ceilings, can it? Is this really what government contractors have to do in order to comply with the regulatory requirements?
Sure, over time the issue will take care of itself, as the proportion of contracts subject to the various (higher) compensation ceilings work themselves out (i.e., you burn off your backlog). But that could take years, depending on the contracts’ various periods of performance. In the meantime, contractors seeking to comply with the cost principle are going to have a challenge, to say the least. It’s going to take additional resources and take additional time to calculate indirect cost rates. It’s going to take additional resources and take additional time to support those calculations through the inevitable audit. It’s going to cost contractors money at a time when budget concerns are in the forefront of the minds of the buying activities. It’s going to cost contractors money at a time in which Congress and Pentagon leaders are trying to reduce the regulatory burden imposed on contractors in order to reduce unnecessary overhead costs and make doing business with the DOD more enticing.
It’s the wrong thing at the wrong time, but there’s nothing to be done about it … is there? Is there anything a contractor can do to comply with the spirit of the cost principle without having to comply with its literal compliance requirements? It would seem not.
It is what it is.
A contractor has to comply and that’s going to be a pain and there’s nothing that can be done about it.
Unless …
Unless DOD comes up with an acceptable approach that minimizes the bureaucratic burdens while permitting contractors to demonstrate compliance with the requirements of the cost principle.
Unless there is some mathematical means of blending all those individual compensation disallowances together, so as to create a single weighted average rate that, when applied to all active contracts, would be acceptable to both contractors and to the DOD.
Oh, wait. There is such a methodology and DOD has endorsed it.
Here’s a link to the official endorsement.
Now the thing is, this DOD-approved methodology is not a panacea. It only applies to DOD contracts and it requires an advance agreement. So contractors may have to have a DOD set of blended rates and a non-DOD set of multiple rates to apply to each civilian agency contract. That still kind of sucks. But if you are a Pentagon contractor and DCAA is your audit agency, this newly approved approach provides an opportunity for you to streamline your rate calculation processes.
But it requires an advance agreement.
We were tempted to opine that the advance agreement is unnecessary, because why would you need an advance agreement to follow the methodology that DOD has already approved? But then we did some research, and we noticed that DCAA does not approve of the “blended rate” approach to excluding unallowable executive compensation. In point of fact, the DCAA CAM (October 2014 edition) states:
Under the blended rate method, the blended rate is applied to both cap-covered and noncovered contract work. This is in violation of Title 31 of the United States Code, section 1301(a), herein referred to as the Purpose Statute, and section 1341(a)(1), herein referred to as the Anti-Deficiency Act. Section 1301(a) (Purpose Statute) requires that appropriations shall be applied to the objects for which the appropriations were made. Section 1341(a)(1) (Anti-Deficiency Act) places limitations on officers or employees of the United States Government expending and obligating amounts exceeding amounts available in the appropriation. Both sections would be violated at most contractor locations since use of a blended rate would result in a predominant misallocation of the unallowable compensation credit to the contract work that is not subject to the cap or authorized by the appropriation.
If contractors do not carry out or expand their blended rate calculation to an appropriate number of decimal places, the impact of the unallowable compensation may not be significant enough to lower the G&A rate. As a result, unallowable compensation costs will not be recovered by the Government.
[CAM Ref. 6-414.9]
Based on the foregoing – and unless DCAA changes its position in response to the official Pentagon policy change – your DCAA auditors are going to question your use of the blended rate methodology and they are going to assert those costs are expressly unallowable. Your executed advance agreement should act to ensure that those questioned costs are not sustained by your cognizant ACO. Consequently, it seems very prudent indeed to execute that advance agreement called-for by the DOD guidance.
Assuming you are going to go forward with the approved blended rate methodology, it is going to be a complicated calculation. As we see it, here are the key steps inherent in the process endorsed by DOD:
-
Establish a list of all employees whose total compensation, as defined by 31.205-6(p), exceeds the compensation cap of $487,000. Determine the aggregate total amount of unallowable compensation in accordance with the cost principle. Hopefully the unallowable compensation will be limited to the G&A expense pool. If not, apportion the unallowable compensation between all affected indirect cost pools.
-
Establish a list of all employees whose total compensation, as defined by 31.205-6(p), exceeds $952.308. Determine the aggregate total amount of unallowable compensation for those individuals. Hopefully you will have less than five such individuals. If not, you will have to develop two pools of unallowable compensation costs: one pool for the “Top 5” individuals and another pool for the total population. Hopefully the unallowable compensation will be limited to the G&A expense pool. If not, see Step 1.
-
Identify all cost-reimbursement contracts that were (or will be) awarded after June 24, 2014, It is expected that the vast majority of those contracts affecting 2014 have already been bid (i.e., they are near-firm backlog). These contracts are subject to the compensation limits you identified in Step 1.
-
Identify all cost-reimbursement contracts awarded prior to June 24, 2014. (These are firm backlog contracts as of that date.) Those contracts are subject to the compensation limits you identified in Step 2. (Remember you may have two groups based on those contracts that are subject to “Top 5” limits and those for which all employees’ compensation must be evaluated.)
-
For all contracts, estimate G&A expense allocation base dollars to be incurred for those contracts. (If another indirect cost pool is affected, you may need to identify the allocation base dollars (e.g., direct labor dollars) of all affected indirect cost pools.) Extend those values out, year-by-year, for the length of those contracts’ periods of performance, as backlog is burned to zero. Adjust for expected contract modifications (e.g., authorized but currently unpriced work). Where necessary, use out-year Business Development forecasts to estimate the G&A allocation base dollars by contract. As before, if other indirect cost pools are affected, then you will have to identify the affected allocation bases of those pools, year-by-year.
-
Aggregate the G&A base dollars by contract groups. Do a check to make sure that, when you add the G&A allocation base dollars together and then look at the G&A allocation base dollars from all other contract types (e.g., FFP or T&M), you get your total forecasted G&A allocation base. If you don’t, then figure out where you went wrong and do the math again. Note that if you have a high percentage of T&M contract work, you may want to add that type to the cost-reimbursable contracts, because the “M” part of the T&M contract may be absorbing a material amount of your allocated G&A expense dollars.)
-
Calculate a ratio of each contract group’s G&A allocation base dollars to the total G&A allocation base dollars. (Do the same for other affected indirect cost pools, if you have any.) Assuming the G&A expense pool is the only pool where you have unallowable executive compensation, you should have three ratios.
-
Ratio of cost-type contracts subject to the high limit, Top 5 population.
-
Ratio of cost-type contracts subject to the high limit, total employee population.
-
Ratio of cost-type contracts subject to the low limit, total employee population.
-
Multiply the ratios you calculated in Step 7 to the appropriate pool of unallowable costs you identified Steps 1 and 2. The results of that multiplication will be your unallowable executive compensation costs.
-
Add the three results together and credit your G&A expense pool.
-
Calculate your DOD G&A expense rate (total claimed G&A expense pool over total G&A allocation base).
Or you can just have three sets of rates and apply them to the appropriate contracts, based on each contract’s effective date. It’s your choice.
But if you are one of those contractors whose ERP system does not easily support multiple billing rate tables, you may have less of a choice than you think.
|
Fixing DCAA
“ The DCAA provides a valuable service to the DoD as their workload permits. One of the major contracting audit services that DCAA provides is to audit contractor's [sic] annual final indirect cost proposals to ensure that the contractor is charging the Government only costs that are allocable, allowable and reasonable. Unfortunately, the demands for DCAA services far outweigh their internal capability and as such, have tremendous backlog for these high risk audits. … Can you make any recommendations to eliminate this imbalance between supply and demand.”
 Why, yes. Thank you for asking. We have several thoughts to offer for your consideration as you ponder this problematic situation.
First of all, we do not agree with your premise that there is an imbalance between supply and demand for DCAA audit services, especially in regard to those services involving audit of contractors’ annual proposals to establish final indirect cost rates pursuant to the requirements of contract clause 52.216-7. There is no imbalance and, until you move from that false foundational premise, you cannot begin to fix the problem.
Until you recognize the real problem you cannot fix it.
Supply isn’t the problem. In 2008 there were 3,500 DCAA auditors. In 2013 there were 4,933 auditors. (According to Chad Braley’s recent presentation at the BDO/PCI DCAA Audit seminar.) Based on those numbers, in five years DCAA added 1,433 auditors. DCAA staffing was increased by more than 40 percent. That whopping increase in the number of auditors didn’t help the situation at all.
In 2008 DCAA issued more than 30,000 audit reports. In 2013 DCAA issued 6,259 audit reports. Over a five year period, DCAA managed to drop its output by nearly a factor of 5 while increasing staffing by 40 percent.
According to DCAA’s Annual Report to Congress, on average it currently takes the audit agency nearly 3 years (1,090 days) to issue an audit report covering a contractor’s proposal to establish final billing rates for a single year. That’s right: it takes DCAA nearly three years to audit one year, on average. That means that (all things being equal) for each year that passes, DCAA falls two years farther behind.
That’s not to say that the audit agency hasn’t made progress in reducing its backlog. It certainly has. But the fact of the matter is that the backlog was reduced by bureaucratic tricks rather than by performing audits. For example, calling a contractor’s “incurred cost proposal” inadequate for audit means that the agency no longer has to count it as backlog. Other contractors have their proposals declared to be “low risk” and are not audited (or have every third proposal audited.)
The foregoing trends are not anomalies; they are facts. The problem of a growing backlog of “incurred cost proposals” awaiting DCAA audit is not a function of the number of auditors. Apparently, the more auditors DCAA has added, the lower the productivity of the average auditor. (This is correlation and not causation, in case anybody cares about such things here.)
DCAA has been adding resources but the problematic situation isn’t getting any better. The situation has gotten to the point where even DCMA representatives making official presentations to the public have stated (in writing) that DCMA is considering implementing innovative processes to finalize contractors’ billing rates in “recognition that business as usual will not allow us to catch up.”
Think about that phrase for a second.
“ Business as usual” will not permit the contracting parties to accomplish the audit, negotiation, and finalization of indirect cost rates in a timely fashion.
“Business as usual” – which means “business conducted in accordance with the FAR and contract terms and conditions” – is no longer seen as an acceptable approach to solve the problematic situation in which the Pentagon and its contractors find themselves.
“Business as usual” worked pretty well for 40 years; but now “business as usual” won’t get the job done, according to DCMA. It seems that even DCMA has realized (finally!) that DCAA’s irrational obsession on compliance with the audit agency’s interpretation of GAGAS has created a problem that needs to be fixed by methods that are other than “business as usual.”
The audit agency paused for a time and rejiggered its audit procedures and reviews to focus on compliance with GAGAS. As a result, audits now take longer. Fewer audit reports are issued. And the backlog remains.
This is not about the “supply” of DCAA personnel available to support audit request. This is about intentional management decisions made by the senior leadership of the audit agency. Nothing more.
Remember that DCAA made an entirely voluntary management decision to stop performing these audits. The audit agency stopped performing them for about two years. The backlog problem was (to a very large extent) intentionally created by leadership decision. And then the new audit procedures only exacerbated the situation. They didn’t fix any existing problems, but they certainly created new problems for all of us to resolve.
DCAA deferred performing “incurred cost proposal” audits and then, when it got around to performing them, performed them with new procedures (including multiple levels of review) that seemed intentionally designed to slow the audit process down. Even though audit reports currently are issued with the speed of molasses poured out of a sealed bottle with a small pinprick of an opening, while buried in ice at the North Pole, those audit reports still lack the necessary quality – at least according to the DOD Inspector General.
DCAA is neither fast nor is it good. And that situation is not the result of a lack of auditors. It is the result of mismanagement.
There is no imbalance between supply and demand. The demand is largely the same. If anything, it has decreased since DCAA now refuses to perform audits of many contractor cost proposals. And DCAA no longer performs adequacy reviews of contractor CASB Disclosure Statements either. Moreover, many non-DOD agencies that used to use DCAA on a reimbursement basis now no longer choose to do so. So it very much seems to us that demand for DCAA’s audit services has (at best) stayed the same, or has even decreased over time.
While demand has stayed flat or trended downwards, the supply of auditors has gone up. So clearly there is no imbalance. That’s not the root cause, so put that red herring out of your mind. Instead, grapple with the notion that the existing resources are sufficient to meet the existing demand, when properly managed.
DCAA has adequate resources. The problem is that the agency has mismanaged the resources it has.
Can the situation be fixed? Doubtful. The parties have come a long way down the road they’re on, and it will be tough to turn around and head in another direction. DCMA seems to be preparing to handle its mission without the assistance of DCAA. The new Director, Ms. Bales, is going to have to make quick and decisive changes if she is to address the fundamental problems facing her agency.
We have offered recommendations regarding how we would attempt to fix the problems at DCAA, were we in charge of the audit agency. We are not going to repeat them here.
But please stop with the notion that there are insufficient auditors to get the work done. There are plenty of auditors. They just need leadership. We hope Ms. Bales will provide it.
DCAA Scrutinizes CAS Cost Impact Proposals
We have long thought that preparing a compliant cost impact proposal is one of the toughest assignments in the field of government contract cost accounting. Generally, a CAS-covered contractor prepares a cost impact proposal when it (1) makes a change from one (compliant) cost accounting practice to another (compliant) cost accounting practice, or (2) has been caught using a cost accounting practice that is not compliant with one of the applicable Cost Accounting Standards, or (3) has been caught using a cost accounting practice that is inconsistent with its established and/or disclosed cost accounting practices.
There are rules governing cost impacts to be found in the CAS regulations themselves. There are rules governing cost impacts to be found in FAR Part 30. There are rules governing cost impacts to be found in one or more of the CAS clauses in a contract. Hopefully, all those rules and regulations and requirements line up nicely so that compliance with one set of rules means compliance with all the sets of rules. But navigating the process is tricky and not for the inexperienced.
The process is made more challenging by the fact that there are actually two types of cost impact analyses: the Gross Dollar Magnitude (GDM) analysis and the Detailed Cost Impact (DCI) analysis. A contractor generates the GDM analysis and submits it to the cognizant Administrative Contracting Officer with the fervent hope that it will suffice. If it suffices, then great. If not, then the contractor has to prepare the DCI analysis, which is not a lot of fun. While the GDM analysis is high-level and generally utilizes backlog as the basis for calculating the cost impacts, the DCI analysis is a contract-by-contract, EAC-by-EAC grind.
Guess which one the DCAA auditors like to see?
Anyway, regardless of one’s feelings about generating CAS cost impact analyses, the fact of the matter is that DCAA has just issued a MRD that is designed to help auditors evaluate the adequacy of the contractor’s submission.
Basically, the cost impact analysis is a proposal that is being submitted to the ACO. We know DCAA already has a Proposal Adequacy Checklist (which was enshrined in the DFARS). Now DCAA wants to apply that same philosophy to CAS cost impacts. Which is nice, we suppose.
The MRD reported that it is deploying a Cost Impact Adequacy Tool to aid auditors in evaluating the adequacy of those cost impact analyses. The MRD stated:
Adequate cost impact proposals are essential to efficient use of audit resources. Prompt adequacy evaluations facilitate the Government’s timely recovery of cost impacts. … The Cost Impact Adequacy Tool lists adequacy criteria, and the regulatory basis, to consider during the proposal assessment. Expand or curtail the adequacy criteria, as necessary, based on the specific circumstances. Audit teams should use judgment and consider whether inadequacies will have a material effect on the proposed cost impact.
We noted that the guidance does not distinguish between a GDM and DCI impact analysis. We hope that the Adequacy Tool makes that distinction. But we don’t know whether it does or not, because the DCAA has not yet made that Tool available to the general public. At the moment, the Adequacy Tool is only available to DCAA auditors.
Because why would DCAA want contractors to know how its cost impact analyses will be evaluated? Heck, that might actually reduce the number of cost impact analyses that DCAA would have to reject as being inadequate!
What happens when DCAA rejects those cost impact analyses as being inadequate?
According to the MRD:
Return proposals that are inadequate for audit to the contractor through the CFAO (Cognizant Federal Agency Official), describing the specific deficiencies. Even if the proposal is not adequate in all respects, the auditor may, at the CFAO’s request, audit and report on the inadequate cost impact proposal to the extent possible. The CFAO could request a non-audit service such as a rough order magnitude (ROM) estimate of the cost impact.
So much fail right there.
First, let’s be clear that it is the ACO who has the authority to accept, or reject, a contractor’s cost impact analysis. This is yet another example of the DCAA usurping authority rightly due to the ACO. Second, about those ROMs. We’ve discussed them before, a time or two. Based on published court decisions and DOD IG reports, they have not been touted as being models of accuracy. In the Pratt & Whitney CAS dispute, the DCAA ROM was more than cut in half over the span of an eighteen month period.
Finally, let’s discuss the Statute of Limitations associated with the Contracts Dispute Act (CDA). Case after case has ruled that the government cannot toll the Statute of Limitations by asserting the contractor’s cost impact analysis was inadequate or inaccurate. That nasty ole six year clock keeps ticking, ticking, ticking away … regardless of what DCAA may assert regarding the adequacy of the analysis. It would seem that DCAA’s propensity to declare anything inadequate for any reason may lead to more missed CDA deadlines, which is very likely the reason for emphasizing the use of auditor judgment to perform audits regardless of minor imperfections in the contractor’s submission.
This is a tough area. It requires a lot of knowledge and some creativity to successfully navigate. Those statements apply to both the contractor and to the auditors. It is also an area in which many complex and costly disputes arise, which is a sad truth that emerges when one or both parties lack the necessary skill sets to successfully navigate the complex process.
|