Wednesday, February 25, 2015

25 Stories About Work - The Difficulty of Verifying Cash

I was recently on a plane doodling and thought of some funny / interesting stories from 25+ years of working and traveling. So I decided to write them up as short, random chapters of a non-book with the title of this post. Hope you enjoy them and / or find them interesting. Certainly the value will be at least equal to the marginal cost of the book (zero)...

The Midwest, early 1990's

For a long time many governmental entities did not have audits from outside firms.  Beginning in the 1980's and 1990's it became common-place for them to have to open up their books and bring in third party professional audit firms to review their accounts.  If they had not been audited before, we called it a "first time through" audit because the amount of work was exponentially higher - you had to document the controls, figure out who was who at the client, validate the opening balances, etc...  Typically after the first audit it was much easier because the 2nd year audit just followed the work papers of the prior year auditors (unless you were like me and asked a lot of questions, which is a story for another "25 stories about work" article).

Recently I thought about my experience when I read this article about the state of New Mexico while reading this article from Bloomberg (a fantastic news source) titled "New Mexico's $100 Million Accounting Error". From the article:
New Mexico can’t balance its checkbook.
Cash in the state’s bank account is at least $100 million short of what’s recorded in the finance department’s ledger, pushing officials to adjust reserves by that amount, to about $650 million. The blame, the current administration says, lies with the introduction of a new accounting system in 2006.
While it would seem astonishing that in this day and age, when you have on-line bank statements and immediate access to data for personal accounts, that a governmental entity could be that far off the mark, it wasn't shocking to me. As a new auditor at this first-time through audit, I was given what was thought to be the simplest of tasks - auditing the cash on the books and reconciling this cash balance to the bank statement.

How you and I and almost everyone else operates is that you have a checkbook balance and as you make a payment (write a check), you deduct that amount from your available cash and you then know how much money you have left in your account. Since deductions can come in many forms (ATM withdrawals, auto-payments, and manual checks) you need to balance your checkbook periodically to make sure you don't miss anything, but other than that it isn't that difficult conceptually. The same process obviously works in reverse for deposits.

The governmental entity I was auditing in the early 1990's, however, used a totally different philosophy. They assumed that they HAD the cash forever until you proved that the check was cashed by whomever they sent the payment out to. Thus when you started to look at the bank balance "on the books", it showed hundreds of millions of dollars. When you looked at bank statement (from the bank), you saw a few million dollars. Thus my nearly insane task was to reconcile out the hundreds of millions of dollars in payments that had been made over the years to get from all the cash deposited back to the few million dollars left on hand. To be fair, staff at the governmental entity had taken a "crack" at this task and there was lots of manual records attempting to bridge the gap, but it was still a giant effort. New Mexico apparently uses the same "model" today - per that Bloomberg article:
Officials commissioned a study on the variances between the state ledger and its bank accounts from fiscal 2007 through February 2013.
Contractors could match only 2 percent of 160 million entries to a corresponding bank transaction, according to a Jan. 19 memo to lawmakers from Legislative Finance Committee staff.
Hundreds of thousands of transactions totaling more than $836 million are absent from the system, the study found. It estimated that the state could have from $76 million to $400 million less than its records reflect. Clifford said he requested $3.4 million to create processes to properly record cash balances. It will take about two years to achieve a “clean” annual financial report, he said. Should the imbalance exceed $100 million, the gap would come out of reserves, he said.
I still remember writing up memos attempting to explain this situation to the partner on the engagement. We did not have a lot of time set aside for auditing cash, which is supposed to be simple, and when you bid out these governmental jobs we were already doing the work at a loss (compared to standard billing rates) so there was little or no tolerance for spending extra work at this unprofitable client. Thus I was not only handed an impossible task my own firm was not pleased with my careful documentation of this situation which caused them to have to spend even more time writing memos to provide credence to the numbers so that we could complete the audit.

I can't speak for New Mexico but the processes I found in the 1990's were unbelievably archaic... they ran checks in a batch and often the checks were mangled in the printer. When this occurred, a staff person put a blank check in the typewriter, typed in all the amounts, and had "their friend" (this was her term, not mine) use the signature stamp on an adjacent desk and then they would send out this manual check instead of the mangled official one. There was literally nothing stopping them from writing out a check to themselves for a lot of money, stamping it with the signature machine on the desk, and then going out and cashing it. Theoretically the "audit" of the cash balances would have found it, but as you can see from the section above auditing out cash took years to do and was a very inaccurate, manual process. When I explained this process to my superior they couldn't believe it (I was "causing trouble" again, just like I did when I documented the insane cash reconciliation process above) but since I carefully documented it they weren't able to refute it. Doing things like this just made me more and more unpopular at the audit, of course.

Another item that I found is that I attempted to reconcile their pension accounts on the books to the statements from the custodians. Of course there were major discrepancies here that I immediately noted; perhaps they had reasonable explanations but at this point in the audit the partner just took those papers away from me and scooted me out. They didn't want me digging into anything else because who knows what I might find. We were being paid poorly in the first place and I was just generating more work; plus it wasn't a good way to win future work by throwing a big new client under the bus (with issues such as this).

You learn about your temperament for a job once you are doing it professionally; I started to learn that I was not good at 1) doing rote tasks in an unquestioning manner 2) accepting things at face value that made no sense to me (even if everyone else didn't have a problem with it) 3) leaving aside problems or potential issues just because they were unfunded, painful, or politically difficult. The audit method back in the early 1990's was just to get the fee, get the job done, check the boxes, and get the heck out of there. I was not a good fit for this model for the reasons listed above (of course, if I was at a well-run company, perhaps I would have been on board with everything, but this never seemed to be the case) and I didn't last too many years in auditing. They were glad to see the last of me, too.

A lot has changed in audits in the 20+ years since this occurred and I assumed that a lot of things have changed for the better and that controls have been upgraded and standards for audits raised. But this New Mexico issue just brings me back to the early 1990's and perhaps my optimism is completely misplaced.

Cross posted at Chicago Boyz

No comments: