4. Developing Pro Forma Fin. Statements, Projecting CF - PowerPoint PPT Presentation

1 / 45
About This Presentation
Title:

4. Developing Pro Forma Fin. Statements, Projecting CF

Description:

Title: 2. The Business Plan Author: richmich Last modified by: richmich Created Date: 8/31/2006 5:20:06 PM Document presentation format: On-screen Show – PowerPoint PPT presentation

Number of Views:150
Avg rating:3.0/5.0
Slides: 46
Provided by: richm178
Learn more at: https://crab.rutgers.edu
Category:

less

Transcript and Presenter's Notes

Title: 4. Developing Pro Forma Fin. Statements, Projecting CF


1
4. Developing Pro Forma Fin. Statements,
Projecting CFs Profits, Assessing Fin. Needs
  • 4.1 Introduction to Pro Forma Financial
    Statements for NVs
  • 4.2 Foundation of New Venture Financial Pro
    Formas Projecting Sales and Revenues Methods
  • 4.2.1 Type of Business
  • 4.2.2 Review of Theory of Demand, the Demand
    Function, Econometric Model of Demand,
    Elasticities
  • 4.2.3 New Product Life Cycle Curve Bass Model
    or the S Curve
  • 4.2.4 Growth Rate Escalation Method
  • 4.2.5 Market Potential and Saturation Method
  • 4.2.6 Estimating Expected Revenues from Large,
    Discrete Contracts

2
4. Developing Pro Forma Fin. Statements,
Projecting CFs Profits, Assessing Fin. Needs
  • 4.3 Projecting Expenses
  • 4.4 Links Between Cash Flows, Net Income, and
    Balance Sheets
  • 4.5 Assessing Financial Investment Needs
  • 4.5 Risk Analysis
  • 4.6 Review of Various New Venture Pro Forma
    Financials

3
4.1 Introduction to Pro Forma Financial
Statements for NVs
  • Pro Forma translated from Latin is for the sake
    of form
  • A pro forma financial statement is a hypothetical
    financial statement that is based on a set of
    assumptions
  • NV pro formas future year projections based on
    little if any history

4
4.1 Introduction to Pro Forma Financial
Statements for NVs
  • Uses of a pro forma for the NV
  • Way to evaluate how much cash the business is
    likely to require and the CF requirements if the
    business grows at a different rate than expected
  • Basis for evaluating and valuating the venture
    and for assessing strategic alternatives
  • Required by VCs for evaluating and valuating the
    venture basis for negotiation of pre- and
    post-money valuations
  • Used as a benchmark to compare actuals v.
    projections feedback for needed changes in
    direction or abandoning the venture

5
4.1 Introduction to Pro Forma Financial
Statements for NVs
  • The most important NV pro forma financial
    statement Statement of Cash Flows
  • The survival of the venture is based on cash, not
    net income
  • Need income statement to develop cash flow
    statement
  • NV business plans usually include income and cash
    flow statements, may or may not include balance
    sheet

6
4.1 Introduction to Pro Forma Financial
Statements for NVs

7
4.2 Foundation of New Venture Financial Pro
Formas Projecting Sales and Revenues Methods
  • The revenue forecast is the driver of the pro
    forma financials and many aspects of the BP.
  • Total Revenue Price (P) x Sales (Qd)
  • Sales and price forecasts drive revenues,
    expenses, and resource needs.
  • Risk analysis is highly driven by sales, price
    and revenue projections.

8
4.2.1 Types of Business
  • Sales, price and revenue projection methods will
    be driven by type of business
  • Mass market mass / database marketing
  • E.g., Website visitation
  • E.g., Retail sales
  • Large contract, client driven Relationship
    marketing
  • E.g., Consulting
  • E.g., Commercial construction
  • E.g., Information systems sales and service

9
4.2.1 Types of Business
  • Sales, price and revenue projection methods will
    be driven by type of business
  • Mass market
  • Sales projections based on percent market
    saturation
  • Product life cycle model such as the Bass model
  • May use an econometric model if have a track
    record
  • Demand equation from microeconomics

10
4.2.1 Types of Business
  • Sales, price and revenue projection methods will
    be driven by type of business
  • Large contract, client driven
  • Sales projections based on the expected value of
    a small number of large, identifiable or generic
    projects / contracts
  • Sales and revenues are usually phased-in yet and
    lumpy (large jumps in revenue as projects
    generate revenues
  • E.g.s Energy services projects (100k-10m, oil
    rig sales (50m - 100m)
  • Price projections based on quantitative value to
    the client and results of negotiations
  • Large leads in cash outflow before cash inflows

11
4.2.2 Review of Theory of Demand, the Demand
Function, Econometric Model of Demand,
Elasticities

Price
Qd
12
4.2.2 Demand Function and Econometric Equation
  • Demand Curve
  • Qd f(P) cet. par.
  • Demand Function
  • Qd f(P, Psubs, Advertising , Income, )
  • where
  • P product price
  • Psubs price of substitutes
  • Income relevant measure of income, such as
    GDP, national income, average customer
    household income,

13
4.2.2 Demand Function and Econometric Equation
  • Econometric Equation of Sales
  • where
  • Pt-1 product price in previous period
  • Psubs,t-1 price of substitutes in previous
    period
  • Incomet-1 relevant measure of income, such as
    GDP, national income, average customer
    household income, in previous period
  • a0, , an regression coefficients
  • X1,t-1, , Xn,t-1 other variables that drive
    sales
  • et error term
  • and expect that a2, a3, a4 are positive and a1
    is negative

14
4.2.2 Demand Function and Econometric Equation
  • Econometric Equation of Sales
  • where the price elasticity of sales (percentage
    change in sales for a percentage change in
    price)

15
4.2.2 Demand Function and Econometric Equation
  • Econometric Equation of Sales
  • where the income elasticity of sales (percentage
    change in sales for a percentage change in
    income)

16
4.2.2 Demand Function and Econometric Equation
  • Elasticities
  • Useful for strategic pricing
  • Price inelastic means that revenues will rise
    with a price increase
  • Income elastic means that sales and revenues are
    very sensitive to the level of economic activity

17
4.2.2 Demand Function and Econometric Equation
  • Econometric Equation of Sales
  • Useful only if secondary information is available
    or can be estimated
  • Secondary information would be the estimation of
    a demand equation based on similar product sales
    and data
  • Estimate for another similar product where the
    data is available and use to project sales and
    strategic decisions
  • Typically used for projections by mass market
    businesses with a history of sales and price
    information

18
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve
  • The marketing research literature has concluded
    that new product life cycle curves typically
    follow an S pattern

19
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve
  • The new product sales model explains this S-curve
    shape based on diffusion theory.
  • Diffusion theory is actually a theory of
    communication regarding how information is
    dispersed within a social system over time.
  • The consumer product adoption process based on
    relative adoption time categorizes individuals as
  • innovators, early adopters, early majority, late
    majority, and laggards.
  • Diffusion theory can be used to develop a
    systematic method for projecting sales that is
    more justifiable then rationalized conjecture
  • Room air conditioners, TVs, VCRs, PCs, cell
    phones to name a few products, all followed a S
    curve form of market saturation

20
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve
  • There are at least three major types of models
    that have been proposed for forecasting new
    product first purchase sales
  • Pure Innovative Models (e.g., Fourt and Woodlock
    1960)
  • Pure Imitative Models (e.g., Fisher and Pry 1971,
    Mansfield 1961)
  • Combination Models (e.g., Bass 1969)

21
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve

22
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve
  • where
  • Qt number of adopters or unit sales at time t
  • p coefficient of innovation, or the likelihood
    that somebody who is not yet using the product
    will start using it because of mass media
    coverage or other external factors Van den
    Bulte (2002)
  • q coefficient of imitation, or the likelihood
    that somebody who is not yet using the product
    will start using it because of "word-of-mouth" or
    other influence from those already using the
    product Van den Bulte (2002)
  • M market size, or ultimate number of adopters
    or unit sales
  • A cumulative number of adopters or unit sales
    to date

23
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve

24
4.2.3 New Product Life Cycle Curve Bass Model
or the S Curve
  • Use the Bass Model
  • Obtain p and q from a similar type of product
    from the marketing literature or estimate them
    for a similar product
  • Estimate own M (or size of market for your
    product) and first year sales or adopters
  • Use M, p, and q to predict Q and A, then apply to
    predicted price to obtain projected revenues

25
4.2.4 Growth Rate Escalation Method
  • Naïve prediction model
  • Assumes growth rate(s) and used to escalate
    sales
  • Based simply on compound growth rates
  • Usually not based on any theory or experience or
    assumption other conjecture
  • Difficult to justify

26
4.2.5 Market Potential and Saturation Method
  • Based on analysis of overall size of market and
    levels of penetration by your venture
  • Similar to S curve as its based on market
    potential
  • Different than S curve as the S curve is based on
    assumed overall size of market for the venture
    only
  • S curve assumed market size (M) should be
    benchmarked to overall size of the market for
    long-term market share

27
4.2.6 Estimating Expected Revenues from Large,
Discrete Contracts
  • Based on knowledge of and relationships with
    targeted potential clients
  • Revenue projection based on expected value of
    revenues generated from each project individually
  • Develop customer sales cycle and typical time
    line then use to project each projects revenues

28
4.2.6 Estimating Expected Revenues from Large,
Discrete Contracts
29
4.2.6 Estimating Expected Revenues from Large,
Discrete Contracts
30
4.3 Projecting Expenses
  • Total Cost (TC) Fixed Cost (FC) Variable Cost
    (VC)
  • In economic theory, FC includes rate of return
    equal to the opportunity cost of capital to the
    equity investors
  • Variable cost items are driven by sales by
    definition
  • Cost of goods sold purely driven by sales
  • Fixed costs items are driven by scale of the
    business and business type
  • Cash Flow Cycle identified employees, materials,
    and fixed assets
  • Capital intensive businesses require high levels
    of fixed assets compared to other factor inputs
  • E.g., consulting firm have little fixed cost
    physical product manufacturing businesses such as
    electric power generation or petroleum extraction
    have very high fixed cost

31
4.3 Projecting Expenses
32
4.3 Projecting Expenses
  • Review microeconomic theory of cost helpful in
    understanding how costs grow and defining a cost
  • Review TC, average TC (ATC), average FC (AFC),
    and average VC (AVC) and marginal cost (MC)
    curves
  • Review long-run average cost curve and economies
    of scale

33
4.3 Projecting Expenses
  • See integrated pro forma financials template for
    typical expense items for a NV
  • List all assumptions to your pro forma revenue
    and expense projections as notes to accompany pro
    formas
  • Pricing (value-based, cost-plus)
  • Method for forecasting sales
  • Timing of fixed asset commitments
  • Terms of loans
  • Salaries by person or title
  • Describe assumptions of any material expense

34
4.3 Projecting Expenses
35
4.4 Links Between Cash Flows, Net Income, and
Balance Sheets
  • The three financial statement are interdependent
    and linked by accounting methods
  • Income Statement changes affect Balance Sheet and
    Cash Flow (e.g., higher profit may lead to
    increased cash balances).
  • Balance Sheet changes affect Income Statement and
    Cash Flow (e.g., borrowing leads to interest
    expense and reduces taxes).
  • An financial model should integrate the
    statements
  • Ideally, pro forma income, cash flow, and balance
    sheet financial statements and sources and uses
    of funds should be developed
  • The most important is the statement of cash flows
    as the NV financial decisions, planning and basis
    for survival are based on cash flow and cash
    needs

36
4.4 Links Between Cash Flows, Net Income, and
Balance Sheets
  • Profitability is not the same as cash flow
  • Non-cash revenues may increase NI but reduce CF
    as credit is extended to the customer
  • Capital acquisitions will decrease CF by the
    amount of the purchase at the time of payment but
    not affect NI
  • Only affect NI when depreciated or expensed
    over its useful life depreciation will affect NI
    but not CF
  • Reflected as a fixed or long-term asset and
    liability is recorded based on how funds obtained
    (e.g. credit long-term debt)
  • See integrated pro forma income, CF, and balance
    sheet templates

37
4.4 Links Between Cash Flows, Net Income, and
Balance Sheets
  • Pro Forma Forecasting No track record to rely on
  • Yardstick Approach
  • Comparable firms in relevant dimensions
  • IPO prospectuses
  • Other data sources
  • Value Line or other investment information
    sources
  • Trade Show Exhibits and Conference Paper
    Presentations
  • Public Record Bids

38
4.5 Assessing Financial Needs
  • Based on the shortfall in cash flow projection
    from pro formas
  • Identifies specific sources of cash
  • CF from operations
  • CF from loans (identify creditors and amounts)
  • CF from equity (identify investors and amounts
    may be a confidentiality or strategy issue here)
  • Identifies uses of cash and schedule of needs

39
(No Transcript)
40
New Company Assumptions
  1. Development will require 18 months, during which
    no sales will be made.
  2. Initial sales of 10,000 in the 19th month.
  3. Sales will grow 8 per month in real terms for
    three years and at the inflation rate thereafter.
  4. Cash operating expenses during the development
    period of 15,000 per month, plus inflation.
  5. Inflation at 9 percent per year.

41
New Company Assumptions
  • A 200,000 production facility will come on line
    at the end of month 18. The facility is to be
    leased by the company for the first 5 years of
    operation, with monthly payments of 3,000.
  • Gross profit of 60 of sales revenue on materials
    costs with trade discounts.
  • Selling expenses of 15 of sales.
  • Administrative expenses of 2,000 per month
    beginning in month 19, growing at the inflation
    rate, plus 15 percent of sales (Included in
    development period operating expense total).

42
New Company Assumptions
  • Entrepreneurs salary of 3,000 per month through
    the first full year of sales. (included in
    initial operating expenses), increasing
    thereafter by 500 per month.
  • Corporate tax rate of 45. No loss carry forward.
  • All sales are for credit. The average collection
    period is 45 days. No discount for prompt
    payment.
  • The inventory turnover rate is 5 times per year,
    measured against ending inventory.

43
New Company Assumptions
  • The company desires to maintain the greater of 30
    days sales in cash or 10,000.
  • All materials are purchased on credit, with terms
    of net 30. The company anticipates paying in
    time to receive the discount. The payables
    period is 10 days.
  • The entrepreneur will borrow any funds necessary
    at a rate of 1 per month.
  • Initial investment by the entrepreneur of
    200,000. Additional financing as needed by
    borrowing on a line of credit.

44
4.5 Risk Analysis
  • Modeling uncertainty is more important for a NV
    than established business
  • Expected projection is an anchor for forecasts of
    uncertainty / scenarios
  • Base on comparable companies
  • Envision alternative realistic scenarios and
    reflect in pro forma projections and estimate
    probabilities
  • Determine 3-4 key risk driving variables and
    re-do projections based on these scenarios
  • Use monte carlo simulation for large NV
    investments

45
4.5 Review Various NV Pro Formas
  • Discuss results from cases
Write a Comment
User Comments (0)
About PowerShow.com