LM07 Company Analysis: Forecasting 2024 Level I Notes

LM07 Company Analysis: Forecasting

1. Introduction ...........................................................................................................................................................2
2. Forecast Objects, Principles, and Approaches ..........................................................................................2
3. Forecasting Revenues ........................................................................................................................................4
4. Forecasting Operating Expenses and Working Capital.........................................................................6
5. Forecasting Capital Investments and Capital Structure .......................................................................8
6. Scenario Analysis .................................................................................................................................................9
Summary................................................................................................................................................................... 12

This document should be read in conjunction with the corresponding learning module in the 2024
Level I CFA® Program curriculum. Some of the graphs, charts, tables, examples, and figures are
copyright 2023, CFA Institute. Reproduced and republished with permission from CFA Institute. All
rights reserved.

Required disclaimer: CFA Institute does not endorse, promote, or warrant the accuracy or quality of
the products or services offered by IFT. CFA Institute, CFA®, and Chartered Financial Analyst® are
trademarks owned by CFA Institute.

Version 1.0

© IFT. All rights reserved 1

LM07 Company Analysis: Forecasting 2024 Level I Notes

1. Introduction
This learning module covers:
• What to forecast, approaches to forecasting, and selecting a forecast horizon.
• Forecasting particular items: revenues; operating expenses and working capital;
capital investments and capital structure.
• Use of scenario analysis in considering multiple outcomes.
2. Forecast Objects, Principles, and Approaches
What to Forecast?
Analysts can focus on different forecast objects such as:
• Drivers of financial statement lines: This concept was covered in the previous learning
module. The net sales for Warehouse Club Inc were analyzed using two bottom-up
drivers: the average number of stores opened and the average net sales per store. Net
sales can be forecasted by forecasting these drivers individually and multiplying them.
The advantage of this approach is that it improves the explanatory value of the
forecast and may also improve accuracy.
• Individual financial statement lines: Instead of forecasting drivers, we can directly
forecast individual financial statement lines. This approach is often used for lines
without clear drivers, for less-material items, and for items that the analyst does not
have a perspective on. For example: amortization expense, analysts may use the
estimate provided by the management or simply assume that the quantity will remain
the same in future periods.
• Summary measures: This includes items like free cash flow, earning per share, and
total assets. The advantage of using these as forecast objects is efficiency. The
disadvantage is reduced transparency, which makes it difficult to audit the forecast.
This method is best suited when the summary measure is stable and predictable, or
when issuer disclosures are severely restricted.
• Ad hoc objects: This includes items that may not have been reported on previous
financial statements. Examples include the outcomes of a significant legal proceeding,
a government regulatory action, a tax dispute, or a natural disaster.
An analyst’s choice of forecast object depends on available information, efficiency, accuracy,
explanatory value, and verifiability.
Focus on Objects That Are Regularly Disclosed
CFAI recommends using forecast objects that are either regularly disclosed or can be directly
calculated using what is regularly disclosed.

© IFT. All rights reserved 2

LM07 Company Analysis: Forecasting 2024 Level I Notes

Non-regularly disclosed information can be used to supplement forecasts but is problematic

for direct use because forecasts cannot be confirmed in a timely manner.
CFAI also cautions against overly complex models, because they require more forecasts and
take more time to update, often with no improvement in accuracy.
Forecast Approaches
For any object there are four general forecast approaches:
• Historical results (assume past is precedent): This approach uses past observed or
calculated values as a forecast. This is an easy to implement approach and may be
appropriate for companies operating in industries where the industry structure is not
expected to change. The approach can also be used to forecast objects that are not
material or that the analyst does not hold an opinion on. The approach is less
appropriate for companies in cyclical industries because the future period could be at
a different point in the business cycle.
• Historical Base Rates and Convergence: This approach uses an industry or peer group
average or median, calculated over a long time period as a “base rate” for forecasting
where an object will converge to over time. This approach is more complex than the
previous approach because analyst discretion is required to select the object, the
sample to calculate the base rate, and a time frame for convergence to the base rate.
The approach may be appropriate for companies operating in well established
industries such as banks, airlines, retailers etc. It is less appropriate for companies in
changing or new industries.
• Management guidance: A company's management may publicly announce earnings,
revenue, and other targets for the next quarter, year, or longer (known as
management guidance or simply guidance). This guidance is often provided as a range
(e.g. sales growth of 3%-5%) and is based on many sub-forecasts and assumptions by
the management about macroeconomic growth, cost inflation, market share changes,
exchange rates etc. Using guidance for forecasts may be appropriate when
management has demonstrated a track record of reliable estimates.
• Analyst’s discretionary forecasts: All other forecast approaches can be grouped under
this category. This can include approaches based on surveys, quantitative models,
probability distributions, and other unobservable inputs. Discretionary forecasts
approaches are commonly used for companies in cyclical industries, companies that
have no comparables, or companies undergoing a fundamental change like a shift in
the competitive or regulatory environment.
An analyst’s choice of forecast approach depends on the company’s industry structure,
sensitivity to the business cycle, and business model, as well as the reliability and availability
of information.

© IFT. All rights reserved 3

LM07 Company Analysis: Forecasting 2024 Level I Notes

Selecting a Forecast Horizon

The forecast time horizon is determined by the investment strategy for which the security is
being considered, the cyclicality of the industry, company-specific factors, and the
preferences of the analyst's employer.
Long-term fund managers may focus their forecasting primarily on the next three to five
years, whereas shorter-term managers may focus more on the next one or two quarters.
3. Forecasting Revenues
Forecast Objects for Revenues
Revenues can be forecasted using a top-down or bottom-up approach. Common top-down
forecast objects include “growth relative to GDP growth” and “market growth and market
share”.
Growth relative to GDP growth approach: In this approach, we first forecast the growth rate
of nominal GDP. We then forecast the company’s revenue growth relative to GDP growth. For
example, we can assume that the company’s revenue will grow at a rate of 150 bps above the
nominal GDP growth rate.
The forecast may also be in relative terms. For example, if we forecast that the GDP will grow
at 4%, and we believe that the company’s revenue will grow at a 20% faster rate, then the
forecasted increase in the company’s revenue is 4% x (1 + 0.20) = 4.8%.
Market growth and market share: In this approach, we combine forecasts of growth in
particular markets with forecasts of a company’s market share. For example, assume Tesla is
expected to maintain a market share of 1% in the automobile market. If the automobile
market is expected to grow to $30 billion in annual revenue, then Tesla’s annual revenue is
forecasted to grow to 1% * $30 billion = $300 million.
Examples of bottom-up drivers for revenue forecast include:
• Volumes and average selling price: Volumes and prices of the company’s products are
forecasted separately and multiplied to get a revenue forecast. For example: The
revenues for a mutual fund can be forecasted based on the AUM and management fee
rates.
• Product-line or segment revenues: Forecasts for individual products, product or
business lines, geographic areas, or reporting segments are made and then aggregated
into a total revenue forecast.
• Capacity-based measure: For example, a retailer’s revenue may be forecasted based on
same-store sales growth and sales related to new stores.
• Return or yield based measure: Forecasts based on balance sheet accounts. For
example, a bank’s interest revenue can be calculated as loans multiplied by the
average interest rate.

© IFT. All rights reserved 4

LM07 Company Analysis: Forecasting 2024 Level I Notes

Separating Recurring and Non-Recurring Revenue or Revenue Growth

While making forecasts, non-recurring items and effects should be excluded. There are two
types of non-recurring items and effects: those that are disclosed by a company’s
management and those that are not, which require analyst judgment to identify.
Examples of the first type include: the effect of changes in exchange rates,
acquisitions/divestitures etc.
An example of the second type is the rapid increase in e-commerce sales during the COVID-
19 pandemic. After things normalized, e-commerce sales as a percentage of retail sales
began to decline.
Forecast Approaches for Revenues
Instructor’s Note: The curriculum illustrates the four approaches to forecasting revenues
through the ‘Warehouse Club Inc’ example. This information is presented as-is from the
curriculum.

© IFT. All rights reserved 5

LM07 Company Analysis: Forecasting 2024 Level I Notes

Analysts must explicitly or implicitly incorporate their view on key risk factors in their
revenue forecasts. Four key risk factors to consider for all companies are competition,
changes in business cycle, inflation and deflation, and technological developments.
4. Forecasting Operating Expenses and Working Capital
Operating costs include cost of goods sold (COGS) and selling, general, and administrative
expenses (SG&A). Disclosures about operating costs are often less detailed than revenue.
Therefore, analysts are often forced to use more aggregated forecasted objects on the
company level (rather than forecasting costs separately for different geographic regions, or
business segments) or summary measures like EBITDA margins. However, forecasts for
operating expenses should still be coherent with revenue forecasts. The choice of forecast
object can vary depending on the forecast horizon.
Cost of Sales and Gross Margins
Since COGS are directly related to sales, we can forecast future COGS as a percentage of
future sales. Analysts should understand the historical relationship between COGS and sales,
and determine if this relationship is likely to decrease, increase, or remain unchanged.
Sales – COGS = gross margin. Therefore, COGS and gross margin are inversely related.
Some factors that analysts should consider while forecasting COGS are:
• Forecasting accuracy can be improved by forecasting COGS for the company’s various
segments or product categories separately.
• Determine if a company has employed hedging strategies to protect its gross margins.
When input prices increase, COGS increase and gross margin decreases. However,
hedging strategies can help mitigate this impact.

© IFT. All rights reserved 6

LM07 Company Analysis: Forecasting 2024 Level I Notes

• Examine gross profit margins of competitors. This can be a useful cross check for
estimating a realistic gross margin. Any difference in gross margins for companies in
the same segment must relate to differences in their business operations.
SG&A Expenses
As compared to COGS, SG&A expenses are less directly related to revenue.
SG&A can be broken down into two components – fixed and variable. The fixed components
such as R&D expenses, management salaries, head office expenses, supporting IT and
administrative operations tend to increase and decrease gradually over time. They do not
fluctuate with sales.
On the other hand, the variable components such as selling and distribution costs are more
directly related to sales. For example, when sales increase, the company may pay out higher
bonuses to its salesforce, it may also hire additional salespersons.
Working Capital Forecasts
Working capital forecasts are typically made using efficiency ratios as the forecast objects,
which are combined with sales and cost forecasts to project accounts receivables,
inventories, accounts payable, and other current asset and liabilities.
Exhibit 9 from the curriculum illustrates the forecast of operating costs and working capital
for Warehouse Club Inc.

© IFT. All rights reserved 7

LM07 Company Analysis: Forecasting 2024 Level I Notes

5. Forecasting Capital Investments and Capital Structure

Forecasts for capital expenditure may differentiate between maintenance and growth capital
expenditures. Forecasts for maintenance capital expenditures are usually based on
depreciation and amortization expenses. Forecasts for growth capital expenditures are
linked to a company's strategy, expansion plans, and revenue growth.
Forecasts for a company’s capital structure take into account the historical leverage ratios
and capital structure, the company’s financial strategy, and capital expenditure forecasts.

© IFT. All rights reserved 8

LM07 Company Analysis: Forecasting 2024 Level I Notes

Exhibit 12 from the curriculum illustrate these approaches for Warehouse Club Inc.

6. Scenario Analysis
The final step in forecasting involves considering the possibility of various outcomes based
on key risk factors and assessing their likelihood of occurrence.
Rather than develop single point estimate forecasts, analysts make several forecast scenarios
that vary based on different outcomes with respect to key risk factors. To make investment

© IFT. All rights reserved 9

LM07 Company Analysis: Forecasting 2024 Level I Notes

decisions, investors compare these scenarios to other analysts' forecasts for a company, as
well as forecasts implied by current valuations.
Technological developments can change the economics of individual companies and entire
industries. Sometimes a technological development results in a new product that threatens
to cannibalize demand for an existing product. For example, introduction of tablets affected
the demand for PCs. To estimate the impact on future demand for the existing product,
analysts forecast the unit sales of the new product and multiply it by an expected
cannibalization factor. The cannibalization factor can be different for different market
segments. For example, the cannibalization factor of tablets is higher in the consumer
segment than the non-consumer segment.
Technological developments can affect the demand for a product, the quantity supplied, or
both. If technological development leads to lower manufacturing costs, then the supply curve
shifts to the right as more quantity is produced at the same price. But if technology results in
better substitute products, then the demand curve will shift to the left.
Example:
(This is based on Example 4 from the curriculum)
The pre-cannibalization PC shipment projections (in thousands) for next year are:
Consumer PC shipments 170,022
Non-consumer PC shipments 180,881
Total PC shipments 350,903
The forecasted tablet shipments (in thousands) for next year are:
Consumer tablet shipments 36,785
Non-consumer tablet shipments 1,686
Total tablet shipments 38,471
Assuming a cannibalization factor of 30% for the consumer segment and 10% for the non-
consumer segment, calculate the post-cannibalization PC projections.
Solution:
Cannibalization in the consumer segment = 0.3 x 36,785 = 11,036
Cannibalization in the non-consumer segment = 0.1 x 1,686 = 169
Therefore, the post-cannibalization PC projections are:
Consumer PC shipments 170,022 – 11,036 = 158,986
Non-consumer PC shipments 180,881 – 169 = 180,712
Total PC shipments 339,698
To evaluate the impact of technological developments on operating costs and margins, we
first need to understand the breakdown between fixed and variable costs. If sales decrease

© IFT. All rights reserved 10

LM07 Company Analysis: Forecasting 2024 Level I Notes

because of cannibalization, then variable costs also decrease. However, the fixed costs
remain unchanged.
Since very few companies provide a breakdown of fixed versus variable costs, we can
estimate them using the following formulas:
% Δ (cost of revenue + operating expense)
% Variable Cost ≈ % Δrevenue
% Fixed Cost = 1 − % Variable cost

© IFT. All rights reserved 11

LM07 Company Analysis: Forecasting 2024 Level I Notes

Summary
LO: Explain principles and approaches to forecasting a company’s financial results and
position.
Analysts can focus on different forecast objects such as:
• Drivers of financial statement lines
• Individual financial statement lines
• Summary measures
• Ad hoc objects
An analyst’s choice of forecast object depends on available information, efficiency, accuracy,
explanatory value, and verifiability.
LO: Explain approaches to forecasting a company’s revenues.
For any forecast object there are four general forecast approaches:
• Historical results
• Historical base rates and convergence
• Management guidance
• Analyst discretion
An analyst’s choice of forecast approach depends on the company’s industry structure,
sensitivity to the business cycle, and business model, as well as the reliability and availability
of information.
Revenue can be forecasted using a top-down or bottom-up approach. Common top-down
forecast objects include “growth relative to GDP growth” and “market growth and market
share”. Bottom-up drivers for revenue forecast include: volumes and average selling price,
product-line or segment revenues, capacity-based measures, return or yield based measures.
LO: Explain approaches to forecasting a company’s operating expenses and working
capital.
Operating costs include cost of goods sold (COGS) and selling, general, and administrative
expenses (SG&A). Disclosures about operating costs are often less detailed than revenue.
Therefore, analysts are often forced to use more aggregated forecasted objects on the
company level (rather than forecasting costs separately for different geographic regions, or
business segments) or summary measures like EBITDA margins. However, forecasts for
operating expenses should still be coherent with revenue forecasts.
Working capital forecasts are typically made using efficiency ratios as the forecast objects,
which are combined with sales and cost forecasts to project accounts receivables,
inventories, accounts payable, and other current asset and liabilities.

© IFT. All rights reserved 12

LM07 Company Analysis: Forecasting 2024 Level I Notes

LO: Explain approaches to forecasting a company’s capital investments and capital

structure.
Forecasts for capital expenditure may differentiate between maintenance and growth capital
expenditures. Forecasts for maintenance capital expenditures are usually based on
depreciation and amortization expenses. Forecasts for growth capital expenditures are
linked to a company's strategy, expansion plans, and revenue growth.
Forecasts for a company’s capital structure take into account the historical leverage ratios
and capital structure, the company’s financial strategy, and capital expenditure forecasts.
LO: Describe the use of scenario analysis in forecasting.
Rather than develop single point estimate forecasts, analysts make several forecast scenarios
that vary based on different outcomes with respect to key risk factors. To make investment
decisions, investors compare these scenarios to other analysts' forecasts for a company, as
well as forecasts implied by current valuations.

© IFT. All rights reserved 13