Fed Rate Cuts Ruin Budgets Everywhere!
By Dave Koch
So, here we are beginning February and already we are preparing to discuss with the board why our budgets are off target. This usually happened in the 2nd or 3rd quarter when we are on our way to achieving our annual results. But after 1 month, what went wrong? For many, this year’s budget had some level of planned rate reductions built in. Did anyone forecast a 1.25% reduction in January? In all my discussions with clients I found no one preparing their board for the effect on budgets from the Fed actions. Isn’t’ the point of running an ALM model to understand the impact of rate movements BEFORE they happen?
This month I wanted to take time to discuss a general approach to budgeting that might prevent a major upheaval in future forecasts and also provide a better measure for our income at risk analysis as well.
Most of you know that we have always preached that budgeting and asset/liability management processes should be using the same assumptions. Many clients’ budgets are built around a flat interest rate forecast as that is the baseline used for income at risk testing. Some clients have branched out in the budgeting process and incorporated a set of forecast of rates as their “base case” prediction. In these cases, the base case is not the basis for income at risk, but is the expected level for budgeted income. Either way, when we get a move like last month, our entire foundation for either analysis is put under pressure.
The question I want to post is whether our prediction of rates, either flat or otherwise, provides us with a blueprint for useful decision making? Note I didn’t ask if the prediction of rates was right, as even the best minds in this area fail to get their projections ‘right” by about 50% of the time. The issue here is what are we using the information for in our decision making process. How can we improve what we use as a major assumption in our modeling to better capture the volatility that exists outside our direct control, but has such a major influence on results?
Usually when we discuss “the budget with a client” we get a response that the budget is the targeted level of income that management is expected to reach. It is not necessarily a collection of plans that will be embarked on to create the flows of balance sheet items that in turn rive income and expense. It is the literal income statement number that is the focal point. If we are to change that reference point, we need to present the budget with the key drivers that cause the result to occur. Income production is predicated on the balance sheet levels that make up the major component of the income statement – interest margin. So when we see the Fed drop rates by 1.25% in a month we get nervous, not only because we know the income will suffer, but because we know that the balance sheet we planned for will now be more difficult to achieve.
For example, if you planned on higher growth in assets, the reduction in rates may cause existing loans to refinance and thus employing loan officer’s efforts to hold existing business. This means that in order to meet the ending balance projections, more volume of loans must be produced. It is like walking on a treadmill, and having someone turn up the speed without warning and trying to keep up with the pace. More flow of volume at lower rates impacts the income levels. Volume drives the income statement as much as rates and the two are linked. So what do you say to in the board meeting about why the income is off?
So, how can we get our process to understand that even our best guess on rates is not a 100% guarantee? Present the budget under multiple rate movements. After all, isn’t that the ALM process for income at risk? When the budget is created there should be a note attached that says “the interest income and expense levels are subject to change based on actual interest rates in the market”. Everyone knows this when they talk about the construction of the budget, but as the variance reports uncover the cause of misses, no one remembers that discussion. So how do we go about incorporating possible rate changes into this process, and how do we document this well?
To make this point, let’s look at a real client budget I was working with last month. This client was adamant about using a flat interest rate level in their budget to take the argument out on what rates might do. My reaction was disbelief. The only rate projection that is 100% wrong is that rates won’t move at all. When pressed, the management team all agreed that the rates would likely fall in 2008 but no one wanted to be the swami that forecasted interest rates. To alleviate this problem, we used a rate forecast service from Global Insight.
Global Insight is one of two offerings we offer that updates rate forecasts for base case, high and low forecast levels. When we ran the new rates against the flat rates the results were amazing. First, when we compared the overall net income levels under all three scenarios, we found that the highest level of income occurred in a flat rate scenario. That means that no matter which way rates moved, they were likely going to miss the budget, and by a good amount in some cases. Look at the decision matrix report below that shows the impact of interest rate movements on projected income levels for each of the 4 rate forecasts we ran.
In comparing net income under these rates we see that income declines when rates fall. The bank is “asset sensitive”. As we look closer at the projections, we see that the Base Case income level is very close to the level projected under the Flat rates. This gives us assurance that as long as rates are close to base we can make our budgeted income. But look at the swings in income. If rates fall faster than anticipated (GI Low) income falls off dramatically. What are the chances that would happen?
By running these number as a part of the budget presentation the board will be expecting that income pressures mount as rates fall. In the ALCO we would discuss if that level of drop is acceptable given our income at risk policy limit, either on the margin or net income. By preparing the budget with rate changes we are linking the budget process to the ALM process as it should be. That still leaves us with the problem of what income level should we post for the budget?
There are many different options here, but for most of you we can simply use a process called “probability weighting”. In our example we will use a rate forecast from Global Insight which offers us both a set of rate movements that are gradual, as well as non-parallel to give us a worst case high and low alternatives as well as a base case. These rates come with a probability assigned to their occurrence. When we run a budget we generally do not expect operating expenses of most fee income lines to be affected by interest rate changes. So, those items which management is able to have direct control over are held constant. The resulting net incomes are then multiplied by the probability for that scenario to arrive at a weighted average value. Below is an example of the previous incomes being weighted for their expected occurrence.
Note that the weighted average $2,670.92 is lower than the base or flat rate levels. This makes sense as there is less than a 100% chance that those rates are right. By presenting this table to the board as a part of the budget, the board becomes more aware of the impact of changes to rates on the performance and can better carry out their fiduciary duty in monitoring management decisions.
So, in presenting budget you have the choice to take the weighted average income levels in as the expected levels, or build the understanding with your board that the Base or Flat levels are not guarantees and that the ALCO process will be forward looking to take changes in any of the key assumptions like interest rate and balance sheet level, and make changes in course to achieve the income target levels in the original budget.
In the end, our goal is to improve our ability to understand the risks in our balance sheet and income statement resulting from the various pressures. Our ALM modeling involves several key assumption areas and the budget review process monthly can be a great source for us to explain the ALM issues to the board as we go. We will examine that reporting and explanation process in our next article. For now, let’s find ways to increase our accuracy in modeling the potential interest rate changes on the income plan we call budgets. Then we can extend that knowledge to our income at risk modeling, by comparing the relative change in earnings to something other than the current use of a flat rate scenario.
To get started in using different interest rates in our forecasts, we offer 2 partnerships to obtain independent rate forecasts for either SAM or FARIN Foresight. The first is the Global Insight rate forecasts. This service updates the rate projections monthly for key interest rates, and provides a written discussion of economic trends and the impact on overall rates. This understanding of how the daily barrage of information on the economy impacts the institution decision making process is a great tool to help boards and others understand the impact key measures have on your world.
The second option comes from McGuire Performance Solutions who offers their Smart Ramp service. The service provides multiple 12 month rate ramp and yield curve shape change scenarios. Financial managers can better estimate earnings at risk IRR and business plan exposures to interest rate changes, facilitating IRR compliance with the least loss to current income and supporting more accurate business plan performance analysis.
For more information on what these services to see samples please contact us at 800-236-3724 and select option 1 for sales. We want to help you to understand your risks and take advantage of the tools you have to make proactive decisions.
Until next time, hold on for more rate cuts…
Dave