Increase operational efficiency
Definition of Cash Reconciliation
Cash reconciliation means comparing internal cash balances against fund accounting and custodians to ensure all systems are in line.
It’s important to distinguish between:
- Trade date cash, which reflects the balances from the day transactions are contractual (e.g. execution of an order)
- Settle date cash, which is the cash sitting in the bank account at the custodian.
If the settlement cycle is 1 day (as for equities in the US), there will be 1 business day lag between trade date cash and settlement date cash.
Cash reconciliation for Hedge funds and Investment Managers
Here are the three providers or systems, whose cash balances should ideally be part of your fund reconciliation:
| Party | Description | Trade date vew | Settle date view |
| Fund admin (or accounting system) | The fund administrator will calculate cash for the NAV, typically per account and currency. | Yes (because it’s what enters the NAV) | No |
| Custodian | The custodian will provide the investment manager with cash statements for each of its accounts – i.e. “cash at bank” information | No | Yes (because it’s what’s actually in your account) |
| Internal system (e.g. Investment Book of Record / IBOR) | The internal system will calculate estimated cash balances. These balances will be for each cash account, per fund, and custodian. | Yes | Sometimes |
Purpose
The primary purpose of cash reconciliation is to ensure internal control and:
- Verify correctness
- Ensure that everything has settled at the custodian
- Verify that the correct numbers have gone into the NAV
- Verify that the investment book of record is calculating cash correctly, so that front office users can trust cash estimates intraday
- Identify issues
- Identify and resolve discrepancies or errors promptly
- Detect and prevent fraud or unauthorised transactions
- Comply with control policies imposed by investors or the board
Or phrased more generally: reduce operational risks.
What cash to reconcile? (balances vs transactions)
Most often, it’s enough to reconcile balances (also called “cash positions”) – and not every transaction. Cash flows can have three sources:
- Fund administrator, e.g. deposits and redemptions.
- The Order Management System (OMS) or IBOR, e.g. trades. If your OMS is also an IBOR, it can be the source of dividends.
- Assets you own, e.g. coupon payments on bonds or dividends on equities.
It doesn’t make much sense for two data sources to send each other data and then compare that they have the same information.
The more helpful check is if the implication of the data sent is the same in both cases, i.e. the calculated cash balances.
Depending on your trading, it could make sense to reconcile transactions, including fees and commissions, before doing the cash reconciliation. The OMS would typically send a trade file to the custodian, which is the basis for settlement. If the information doesn’t match with the counterparty, trade settlement will fail. This is why using a matching platform or doing transaction reconciliation can be a good idea, especially if trade volumes are high.
By reconciling trades, you can find minor differences in commissions, taxes, other fees or even different trade dates or settlement dates before you reconcile the cash balances. It’s much harder to understand why there is a difference when comparing two balances than when comparing trade-level data.
Key Steps in Cash Reconciliation
- Import data
Collect cash balance data from internal systems. The data is already there if reconciliation is part of your investment management system. Ensure data can be automatically imported from the custodian and fund administrator, ideally without being dependent on the vendor for such import and without having to write code (see Limina’s approach to data import here) - Aggregate
If data is already aggregated, e.g. by currency, in one source – you must aggregate the other source to compare at the same aggregation level. More on this below. This aggregation should be fully automated. - Reconcile
Match cash balances. The reconciliation tool will compare line by line, marking matching cash balances and flagging unmatched. Reasons for breaks can be value differences (within pre-set tolerances) or missing lines. - Investigate Discrepancies & Resolve Breaks
The reasons for unmatched cash balances or discrepancies are investigated. The investigation may involve communicating with front office users internally, custodians, brokers, or fund administrators. Depending on the reasons uncovered, the data in the IBOR is either updated or changes made at external parties should the error reside there. If the reconciliation engine is part of the IBOR, updating data can easily be fully automated (as is the case with Limina). - Documenting for Audit
All breaks and actions taken need to be documented. Such documentation is typically created automatically in a reconciliation tool.
Note: There is no audit with Excel, so you must manage documentation manually if you use Excel for your investment reconciliation process.
Frequency of cash reconciliations
A cash recon is best performed at least as often as your NAV is struck. If your fund is daily traded, a daily (or even intraday) reconciliation is the best choice.
Side note on other types of recs: NAV reconciliation isn’t always required, and some firms rely on a combination of P&L-, trade-, cash- and position reconciliation as enough to claim shadow NAV.
Common Challenges in Cash Reconciliation
Future Trends in Cash Reconciliation
Let’s explore some of the emerging trends and shifts in industry practices shaping the future of cash reconciliation.
Artificial Intelligence (AI) and Machine Learning (ML) in Cash Reconciliation
Artificial intelligence (AI) and machine learning (ML) technologies have already started to power cash reconciliation processes. AI-powered solutions can learn from historical break reasons and suggest the most likely reasons for breaks.
AI-powered solutions are, at the same time, not suited for automatic identification of discrepancies. This is because AIs are statistical models, so there is always a risk the AI will make incorrect decisions. In processes where the tolerance for error is zero, such as reconciliation, AI is therefore not suited for full automation but rather for highlighting recommendations to humans.
Outsourcing
Some investment firms have outsourced reconciliation to external firms specialising in cash reconciliations and break resolution. This can be either specifically for reconciliation or, more broadly, for middle office operations functions where cash recon is just one of the services provided.
Intraday reconciliation
Another shift is the move towards real-time or near real-time reconciliation. Traditional reconciliation processes often rely on end-of-day or batch processing, leading to delays in identifying and resolving discrepancies. As some fund admins and custodians have started to offer real-time data feeds and advanced reconciliation systems, investment managers can now perform continuous reconciliations throughout the day, enabling quicker detection and resolution of issues.
