wcm ganpat

Upload: parth-prajapati

Post on 10-Apr-2018

252 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/8/2019 WCM Ganpat

    1/32

  • 8/8/2019 WCM Ganpat

    2/32

    Cash Management

    Cash management is concerned

    with the managing of: cash flows into and out of the firm,

    cash flows within the firm, and

    cash balances held by the firm at a

    point of time by financing deficit or

    investing surplus cash

  • 8/8/2019 WCM Ganpat

    3/32

    FourFacets of Cash Management Cash planning

    Managing the cash flows Optimum cash level

    Investing surplus cash

  • 8/8/2019 WCM Ganpat

    4/32

    Optimum Cash Balance Optimum Cash Balance under Certainty:

    Baumols Model, Beranek Model

    Optimum Cash Balance under

    Uncertainty: The MillerOrr Model

    And Stone Model

  • 8/8/2019 WCM Ganpat

    5/32

    The Baumol Model

    Assumptions: The firm is able to forecast its cash needs

    with certainty.

    The firms cash payments occur uniformlyover a period of time.

    The opportunity cost of holding cash is

    known and it does not change over time. The firm will incur the same transaction cost

    whenever it converts securities to cash.

  • 8/8/2019 WCM Ganpat

    6/32

    The Baumol Model

    The firm incurs a holding cost for keeping the cash balance. It is anopportunity cost; that is, the return foregone on the marketablesecurities. If the opportunity cost is k, then the firms holding cost formaintaining an average cash balance is as follows:

    The firm incurs a transaction cost whenever it converts its marketablesecurities to cash. Total number of transactions during the year will betotal funds requirement, T, divided by the cash balance, C, i.e., T/C. The

    per transaction cost is assumed to be constant. If per transaction cost isc, then the total transaction cost will be:

    The total annual cost of the demand for cash will be:

    The optimum cash balance, C*, is obtained when the total cost isminimum. The formula for the optimum cash balance is as follows:

    Holding cost = ( / 2)k C

    Transaction cost = ( / )c T C

    * 2cTC

    k!

    Total cost = ( / 2) ( / )k C c T C

  • 8/8/2019 WCM Ganpat

    7/32

    IllustrationBaumol Model

    Advani hemical Limited estimates its total cash requirement as Rs 2 crore next year. The companys opportunity

    cost of funds is 15% per annum. The company will have to incur Rs 150 per transaction when it converts its short-termsecurities to cash. Determine the optimum cash balance. How much is the total annual cost of the demand for the

    optimum cash balance? How many deposits will have to be made during the year?

    T!

    000,200Rs15.0

    )000,000,20)(150(2C* !!

    The annual cost will be:

    3=+=+=

    /+/=co total

    During the year, the company will have to make 100 deposits, i.e. converting marketable securities to cash.

  • 8/8/2019 WCM Ganpat

    8/32

    The Beranek Model

    This Model assume that the cash inflowsare steady but the outflows are periodic

    This is the mirror image of the time patternof cash flows in Baumol model, whereinflows are periodic and outflows aresteady

    Cash would be collected continuously at ata uniform rate, but would be disbursedover a short period of time

  • 8/8/2019 WCM Ganpat

    9/32

    In this pattern of cash flows, the challenge

    is to profitably invest the funds between

    the time of their receipts and the when a

    group of cheques are presented to the bank

    for the payment.

    The trade off between interest income andtransaction cost and strategies employed

    are parallel to those of Baumol Model

    The Beranek Model

  • 8/8/2019 WCM Ganpat

    10/32

    The formula for optimum number of

    transaction is same as that for the Baumol

    Model

    However, the cash is accumulated

    gradually (rather than disbursed gradually),

    so the transaction pattern would involve a

    series of cash investment followed by one

    disinvestment at the end of the period.

    The Beranek Model

  • 8/8/2019 WCM Ganpat

    11/32

    The MillerOrr Model

    The MO model provides for two control limitstheupper control limit and the lower control limit as wellas a return point.

    If the firms cash flows fluctuate randomly and hit theupper limit, then it buys sufficient marketablesecurities to come back to a normal level of cashbalance (the return point).

    Similarly, when the firms cash flows wander and hitthe lower limit, it sells sufficient marketablesecurities to bring the cash balance back to the normallevel (the return point).

  • 8/8/2019 WCM Ganpat

    12/32

    The Miller-Orr Model

    The difference between the upper limit and the lower

    limit depends on the following factors: the transaction cost (c)

    the interest rate, (i)

    the standard deviation (W) of net cash flows.

    The formula for determining the distance between upper

    and lower control limits (called Z) is as follows:1/ 3(Upper Limit Lower Limit) (3/ 4 Transaction ost ash Flow Variance/ Interest Rate)

    Upper Limit Lo wer Limit + 3

    Return Point Lo wer Limit +

    The net effect is that the firms hold the average the cash balance equal to:

    verage

    Z

    Z

    ash alance Lower Limit + 4/3Z

  • 8/8/2019 WCM Ganpat

    13/32

    3bW 2

    RP = 3 +LL

    4I

    UL = 3RP2LL

    where: RP = return point

    b = fixed cost per order for converting

    marketable securities into cash.

    I = daily interest rate earned on marketable securities

    W2 = variance of daily changes in the expected cash balance

    LL = the lower control limit

    UL = the upper control limit

    The Miller-Orr Model

  • 8/8/2019 WCM Ganpat

    14/32

    aja td. provide the following information

    Annual yield: 12%, daily: 12/360= 0.0333%

    Transaction cost: s. 1600

    Change in daily cash balance, S.D., s.5000 Minimum Cash balance: s. 50000

    Illustration: Miller-Orr Model

    3bW 2 3*(1600)*(5000) 2

    RP = 3 +LL = 3 +50000

    4I 4 * (0.00033)

    UL = 3RP2LL

  • 8/8/2019 WCM Ganpat

    15/32

    This model is based on the argument that the average

    bank balance is determined by banks requirement of

    maintaining compensatory bank balance in current

    account when an overdraft is taken. (This practice isno more in India)

    Minimizing transaction cost subject to the constraints

    of matching average balance with target balance

    se control limits similar to Miller and Orr Model

    Except

    Two sets of limits : outer limit and inner limit

    The Stone Model

  • 8/8/2019 WCM Ganpat

    16/32

    When the balance reaches the outer limit, the finance

    manager does not trigger securities purchase, as in

    the case of Miller, but checks with the forecast for

    the next pre-determined days. If he finds that the

    forecasted cash flow is expected to move within the

    inner limits, i.e. the balance is closer to target

    balance, he does not order any securities transactions

    and hence save on transaction cost. If forecast is out side the inner limits, he orders

    buying or selling securities to reach the target

    balance

    The Stone Model

  • 8/8/2019 WCM Ganpat

    17/32

    A company has following forecast of cash flow during the next14 days ( s. in akh)

    Illustration: The Stone Model

    Day 1 2 3 4 5 6 7 8 9 10 11 12 13 14Cash

    Forecast 8 4 -5 -4 5 -1 2 0 1 -16 7 -5 2 -5Cash

    Actual8 3 -6 -3 6 -2 5 -4 0 -18 6 -6 -3 4

    The finance Manager has set up upper outer limit of s. 40 lakh

    and the lower outer limit of s. 23 lakh. Inner limit fixed at s.2lakh minus or plus respectivly from the outer limits. i.e. 40-2=38and 23 2 = 25 lakh

    ook out time is 3 days

    Day opened at target cash balance of s. 30 lakh

  • 8/8/2019 WCM Ganpat

    18/32

    Time Pattern ofCash flow

    Baumol Model Beranek Model

    All cash flows are certain All cash flows are certain

    Cash inflows are periodic

    and instantaneous

    Cash outflows are periodic

    and instantaneous

    Cash outflows occur at a

    constant rate

    Cash inflows occur at a

    constant rate

  • 8/8/2019 WCM Ganpat

    19/32

    Time Pattern ofCash flow

    Miller-Orr Model Stone Model

    Firm has minimum required

    cash balance

    Firm has minimum required

    cash balanceCash flows are normally

    distributed and standard

    deviation does not change

    Firm has some knowledge

    of future cash flows

    The Expected cash flow is

    zero

    Although this knowledge

    contain error component

    There is no autocorrelation

    in cash flows

  • 8/8/2019 WCM Ganpat

    20/32

    ReceivableManagement

  • 8/8/2019 WCM Ganpat

    21/32

    Heuristic Approach to

    Establishing Credit imit Heuristic approach is based on companys

    actual experience

    The formula or procedure here has the

    weightage of managerial experience and

    intuition behind it and therefore it is heuristic

    in nature To establish a credit limit and granting credit,

    following factors need to be considered

  • 8/8/2019 WCM Ganpat

    22/32

    Contributing

    Factor

    Rating Credit limit

    %of net worth

    Applicants Credit

    Requirements

    (C)

    C

  • 8/8/2019 WCM Ganpat

    23/32

    Contributing

    Factor

    Rating Credit limit

    %of net worth

    Profit Margin

    (M)

    M

  • 8/8/2019 WCM Ganpat

    24/32

    Sequential Decision Analysis In this analysis, investigation is carried out

    further stage only if the benefit of such

    analysis outweighs its cost

    Required to prepare decision tree

  • 8/8/2019 WCM Ganpat

    25/32

    Discriminant analysis is a statistical tool helpful for classificationpurposes.

    To convey a basic understanding of discriminant analysis, we shallemploy the following assumptions:

    There are two discrete groups (group 1 and group 2).

    Two variables combined in a linear manner would be used for

    discriminating between the two discrete groups (This meansthat the discriminant function will be Zi= aXi+ bYi).

    These two variables arise from multivariate normalpopulations. While the means of the two variables in eachgroup are different, their variance/covariance matrix isidentical for each group.

    Discriminant analysis and

    customer classification

  • 8/8/2019 WCM Ganpat

    26/32

    1. Estimation of the discriminant function

    2. Choice of the cutoff point for the discriminant

    function

    3. Examination of the predictive ability of the

    discriminant functionContd.,

    Steps involved in Discriminant

    analysis

  • 8/8/2019 WCM Ganpat

    27/32

  • 8/8/2019 WCM Ganpat

    28/32

    In n n n num ntsthe et een

    up sum fsquares fZiscoresandthedenom natorrepresentsthe

    thin roupsum ofsquaresoftheZi scores.For maximising the ratio, given in quation, e set its partial

    derivatives ith respect to a and (the t o parameters of thediscriminant function) equal to 0. doing this, e get the follo ing

    normalequations:

    a . 2x b. xy x

    a . xy b

    2y y

    here 2x varianceof X

    xy =covarianceofXand

    2y=varianceofYx=difference et eenthe meanvaluesofXforthet ogroupsy=difference et eenthe meanvaluesofY forthet ogroups.

    Contd.,

  • 8/8/2019 WCM Ganpat

    29/32

    Solving the above normal e uations, we get

    2y. dx - xy. dy

    a =

    2x

    . 2y - xy. xy

    2x. dy - xy. dx

    b =

    2x. 2y - xy. xy

    Contd.,

  • 8/8/2019 WCM Ganpat

    30/32

    The cutoff value which is supposed to separate the two groups, should be chosen in such a

    way that it minimises the probability of misclassification. In theory, the cutoff value isrepresented by the point Z* because it results in minimal misclassification.

    Discriminant functionZ1 Z* Z2

    Group 1 Group 2

    Contd.,

    ProbabilityDistributions and Cut-off Level

    Choice of the cut-off point for the

    discriminant function

  • 8/8/2019 WCM Ganpat

    31/32

    In practice, the cut-off point is chosen in such a way that it

    minimises misclassification in the estimation sample. Todefine this cut-off value, we proceed as follows:

    (i) Calculate the Zi value for the observations and arrange

    them in an ascending order.

    (ii) Consider the mid-points of adjacently ranked observations,

    in the area where the observations from the two groups

    overlap, as possible cut-off levels.

    (iii) Choose that cut-off point which minimises the total number

    of misclassifications.

  • 8/8/2019 WCM Ganpat

    32/32

    Hence , it is desirable t examine the redictive abilit f the

    discriminant f ncti n with reference t a new samp le which may be referred t as a va lidati n r holdout samp le . This invo lve s comparing

    the classification of observa tions based on their discriminant function

    scores with their actual classification. or this purpose the following

    classificationma trix (confusionma trix) is prepared:

    Actu l l ssification

    Group 1 Group

    Pre icte Group 1 11 12

    lassification Group 2 21 22

    If the discriminant mode l correctly predicts the group to whicheachobserva tionbelongs, all observa tions will be on the ma indiagonal

    of the ma trix , i.e . in the C11and C22elements. The percentage of correct

    predictions is:

    C11 + C22

    C11 + C22 + C21 + C22

    Examination of the predictive ability of the

    discriminant model