In the midst of my classes with Eric, I continued to have core accounting classes; I continued to be steeped deeper and deeper in the accounting tradition of information gathering and reporting. In these classes I learned about things called subsidiary ledgers, places where particular types of accounting entries are kept. These ledgers helped simplify the accounting process by not requiring the recording of both sides of the journal entries, and recording other items of interest for reporting.
Accounting has a reputation for being inflexible and unbending in the way it is done; creative accounting is something no ethical person aspires to. But that reputation isn’t accurate. The number of ways to record the information and produce the reports is very diverse, even within the double-entry system. And if subsidiary ledgers and other ways of recording information are taken into account, the options are even more diverse. Understanding an alternative can help make the REA theory clearer. One of the most common alternative accounting approaches is called singled sided accounting. The use of business events might be confused for this form of accounting. Let’s make it clear what this is.
Single Sided Accounting
Single sided accounting as an alternative form of the double-entry accounting system has been around perhaps as long as or longer than the double-entry system and elements of it are very common even in today’s financial reporting systems. Take for example the check register in our personal financial system. There is no reason to make two rows to record each check that is written, (1) one that says what liability was decreased or expense incurred, and (2) the other to show the decrease in cash in the checking account. The reason is that almost all the cash side entries would look exactly the same. By definition, recording the expense or liability side of the entries in the checking account register means we can infer what are the credit or debit entries to cash.1
These cash rows are called the “offset”; they are just the opposite side and always to the same account. The single sided accounting entries in the check register record our cash transactions and we only make the offset entry to the actual cash account periodically. We might record the offset to the general ledger when we reconcile the checking account monthly to reflect the cumulative effect of the changes in cash. In effect for the double-entry system, we have one massive journal entry posted monthly, where all the things we purchased are one set of rows, all the causes of deposits like our paychecks are the other rows, and the difference between the two is a single row offset increase or decrease in cash.
Now if every transaction we did always involved cash in the checking account, we could make our personal financial system such that we only record half of the rows in all of our entries. But this is too simplistic even for our simple example because we also have credit cards; we need to record how we paid because sometimes we used cash, other times the credit card.
Eric taught us that the “R” in REA stands for resource. If we were to view the cash or credit cards as a resource, would that change our system? In other words, can we make up a computerized financial system that records the events, and uses resources as a way to get to the financial statements?
Let’s see if this system might work. We could decide that:
- Each Balance Sheet account is a Financial Resource
- Financial Resources are used to pay for or accomplish any Financial Event
- Financial Events are Income Statement accounts
- Financial Resources can be transferred from one type to another, thus cash can be used to pay the credit card balance.
Creating such examples seems to have been a cottage industry for those interested in McCarthy’s theory in those days; many examples were probably much more elaborate and comprehensive. I believe Eric’s doctoral thesis was one example. This approach can be described quite easily to an accountant by noting that almost any Balance Sheet account can be calculated from the Income Statement rows alone. I remember Eric suggesting if we wanted to know what inventory was, we could add up all the goods produced, subtract goods shipped, and the difference are goods on hand. Want to know accounts receivable? Add up all the sales, subtract all the receipts and the difference is accounts receivable.4
My simple example is meant to show that an alternative to the traditional accounting model can work; it is not a serious suggestion of the “right” way to build an accounting system. It shows that a system which does not much resemble a double-entry accounting system could be considered the book of record. The primary benefit of my simple system would be many fewer recorded rows. Again, this isn’t a big improvement; but perhaps important as we will see later on. Let’s see if this system can be extended to make it even more useful.
The A in REA stands for Agents. Eric taught us that the definition of Agents was the suggested approach to deal with the “who” of accounting.
Cost centers, as discussed in the last chapter, record who inside the organization was responsible for a particular financial transaction. Adding cost center had a ripple effect upon our financial system, requiring either a new ledger or additional types of entries, or new trial balances to produce the different types of reports expected.
Our checking account register is an example subsidiary ledger in which we only make one sided entries, and then periodically make a two sided entry to the general ledger to record the net effect upon cash for a lot of transactions. Early on, well before computers, accountants realized that because only one side of the entry is made in the subsidiary ledger, adding a column to those journals didn’t have the same ripple effect through the general ledger. Thus subsidiary ledgers are used to store additional information that is only of limited interest.
McCarthy recognized that a great deal of effort is also spent on recording the external “who” that is involved in a transaction; the “who” beyond that recorded in cost center. Most of this complexity in financial systems is embedded in subsidiary ledgers as well.
Most businesses have at least two types of subsidiary ledgers, accounts receivable and accounts payable, and two types of external agents: customers and vendors. Businesses track how much customers owe in the accounts receivable subsidiary ledger, and how much they owe vendors in the accounts payable subsidiary ledger. The one sided journals used to record these activities and the corresponding subsidiary ledgers are very similar for each.
Each customer or vendor is typically assigned a unique ID. If customers or vendors can have multiple accounts with our business (if we are a bank, one customer might have a checking account and a savings account), we might also have an account number (Note: not a GL account, but a customer or vendor account) recorded at the same time on the subsidiary ledger.
Just like the relationship of the general journal entries to the general ledger, the subsidiary journals are posted to the subsidiary ledger. Thus at any point in time, a particular customer or vendor balance can be found.5
This subsidiary journal and ledger approach to the reporting problem has the same benefits and draw backs as the general ledger. The balances are stored to preserve the investment in arithmetic, but we now have multiple possible parts to the book of record, each with some portion of the information we recorded at the time the business event happened.
Alternative Agent Approach
Alternatively, we could eliminate the need for sub-ledgers if we add another two columns to the cost center example above, one for internal agent and another for external agent. We could then use the computer to add up the rows by the columns of interest to produce the report that we want when we want it. This approach shows that the REA model not only makes marginal improvements on the existing reports, such as allowing production of an income statement at any point in time and reducing the number of entries required to produce the report, but it significantly reduces the number of ledgers maintained and intermediate journals needed to post to each of these systems. This change in definition of the book of record has far reaching impacts for how reporting systems are constructed.
The benefits of making this change alone may not outweigh the cost. This approach requires keeping the journal entries from the beginning of time to calculate right now any report we need. So the reduction in the number of systems and subsidiary ledgers may be outweighed by the cost of the engineering needed to make the computer capable of summarizing all that data even if it were kept.
Eric, and the REA theory, suggested there is actually one more possible definition of the book of record: the actual receipts, bank and other statements used to create the journal entries. The real power of the theory is demonstrated when we expand the accounting model to include new types of information. To do that, we will need to analyze our information needs in greater detail. The steps we will follow came in a subsequent class.
Previous: Chapter 6: Business Events
Parent Topic: Part 2: The Professor