Chapter 8 — Index Numbers

Definition: An index number is a single figure that shows, in simple form, the relative change in a variable (or a group of related variables) over time or between places, taking a chosen period (or place) as the base. By convention the value of the index in the base period is 100.

Meaning and purpose:

• Index numbers summarize changes in price, quantity or value of a group of items over time (e.g., prices this year compared with a base year).
• They convert many observations into a single understandable measure to show the magnitude and direction of change (increase, decrease or stability).
• Common uses: measuring inflation (CPI), wholesale price movements (WPI), cost of living, and performance of stock markets (e.g., Sensex, Nifty).

Key characteristics:

• Relative measure: shows change relative to base period (base = 100).
• Dimensionless: has no physical unit—expressed as an index number or percentage.
• Summarizing device: reduces a large set of data to a single comparable figure.
• Depends on choice of base period, items included and weights used.

Types (brief): Price index, quantity index, value index, simple (unweighted) and weighted indices. Weighted indices are preferred when items have different importance (weights).

How to construct (basic steps):

1. Select the items (market basket) whose change is to be measured.
2. Choose a base period and set its index = 100.
3. Collect prices/quantities for base and current periods.
4. Decide whether weights (importance) are needed and choose appropriate formula.
5. Compute the index and interpret (index >100 means increase, <100 means decrease).

Limitations (brief):

• Choice of base period and items can bias results.
• Quality changes, new products and substitution by consumers are hard to capture.
• Aggregation problems—combining heterogeneous items may hide individual movements.

Simple numeric illustration: If price of an item was 50 in the base year and 60 now, the price relative = (60/50)×100 = 120. This means prices rose by 20% compared to the base year.

Index numbers are statistical measures that show relative change in a variable or a group of related variables over time or across regions, expressed with respect to a base value (usually taken as 100). The "Uses and Importance" of index numbers arise from their ability to simplify complex changes into a single comparable figure and to convert nominal measures into real measures. They are indispensable for economic analysis, policy-making and business decisions.

Key uses and reasons why index numbers are important:

• Measuring Inflation and Price Level Changes: Consumer Price Index (CPI) and Wholesale Price Index (WPI) track inflation. Policymakers use them to monitor inflationary trends and design monetary policy.
• Cost-of-Living Adjustments (COLA): Wages, pensions, rent, and government benefits are often indexed to price indices to preserve real purchasing power.
• Deflating Nominal Series: Converting nominal GDP, incomes, wages, or revenues into real terms by dividing by a price index (to remove price effects), enabling meaningful time comparisons.
• Comparing Regions or Time Periods: Index numbers simplify comparison (e.g., price levels in two states, or cost of living across years) using a common base.
• Policy Formulation and Evaluation: Governments use indices to set interest rates, tax brackets, subsidy levels and to evaluate the impact of policies on prices and living standards.
• Business Planning and Cost Control: Businesses use indices to forecast costs (raw materials, wages), set prices, prepare budgets and adjust long-term contracts.
• Measuring Productivity and Competitiveness: Price indices for inputs and outputs (e.g., export and import price indices) help compute real productivity and competitiveness indicators.
• Index-based Financial Instruments: Inflation-indexed bonds, pensions and derivatives use indices for payments and valuation.
• Simplifying Complex Data: A single index summarizes movements in many items (a basket of goods), making communication and interpretation easier for analysts and the public.

Important cautions: index numbers depend on the choice of base period, the composition and weights of the basket, and the formula used (simple average, Laspeyres, Paasche, Fisher). Different choices can yield different numerical values, so interpretation should consider methodology.

Definition: An index number is a statistical measure showing relative change in a variable or group of related variables with respect to a base value taken as 100. Types of index numbers classify indices by purpose, method of construction and choice of base period.

1. By purpose (what is being measured)

• Price Index Numbers – measure changes in prices (for example, CPI, WPI). They show how the price level of a basket of goods changes over time.
• Quantity (Volume) Index Numbers – measure changes in physical quantities produced, consumed or sold.
• Value Index Numbers – measure changes in total monetary value (price × quantity). They reflect both price and quantity changes.

2. By construction method

• Simple (Unweighted) Index Numbers – all items are given equal importance. Examples: simple aggregate index and simple average of price relatives.
• Weighted (Composite) Index Numbers – items are given different importance using weights (usually quantities or expenditures). Important weighted indices are:
• Laspeyres Index – uses base-period quantities as weights (good for cost-of-living measured with a fixed basket).
• Paasche Index – uses current-period quantities as weights (reflects current consumption pattern).
• Fisher Ideal Index – geometric mean of Laspeyres and Paasche; often called the ‘ideal’ index because it reduces bias.
• Other forms – Marshall-Edgeworth (average of quantities as weights) and Drobisch-Bowley variants – less commonly used at basic level.

3. By time-base

• Fixed-base Index – all comparisons are made with one fixed base period (index value in base = 100).
• Chain-base Index – successive short-period indices are linked together to produce a long-run index; useful when baskets or weights change frequently.

Key ideas and advantages/limitations

• Price indices are used to measure inflation and cost-of-living changes. Quantity indices measure real growth in production or consumption.
• Simple indices are easy to compute but ignore relative importance of items. Weighted indices are more accurate but require up-to-date weights.
• Laspeyres tends to overstate price increases (because it uses fixed base quantities), Paasche tends to understate (because it uses current quantities), Fisher balances the two.
• Choice of base period and basket affects results; chain indices help adapt to changing consumption patterns.

Variables used: p0, q0 = price and quantity in base period; p1, q1 = price and quantity in current period (or period 1).

What is a base year? The base year is a reference year against which values in other years are compared when constructing an index number (for example a price index). By convention the index for the base year is set to 100. All other years' index values show percentage change relative to that base year.

Index format (base = 100) An index number in the usual format is:

• Index in year t = (Value in year t / Value in base year) × 100
• So if index(base year) = 100, an index of 120 means a 20% rise versus base year; an index of 85 means a 15% fall.

Kinds of base: Fixed-base and chain-base.

• Fixed-base index: All years are compared to a single fixed base year (e.g. 2010 = 100).
• Chain-base (or linked) index: Each year is compared to the previous year and the links are multiplied to get a continuous series. Chain indices are updated frequently and reduce distortion when structure changes fast.

Why choose or change a base year? The base year should be representative (normal economic conditions), recent enough to reflect consumption/production patterns, and have reliable data. Statistical agencies periodically rebase (change the base year) to reflect structural and consumption changes and new expenditure weights.

Rebasing and conversion — when the base year is changed you convert old indices to the new base using the index of the new base (expressed in the old base):

• NewIndex = (OldIndex / OldIndexOfNewBase) × 100

Important points:

• Setting base year index = 100 makes interpretation easy (percentage change).
• For price indices that use weights, the choice of base affects the weights (quantities/expenditures) used in aggregation.
• When comparing indices with different bases, always convert to the same base before comparing.

Definition: A price relative for an item is the ratio of its price in the current (comparison) period to its price in the base period, expressed as a number or percentage. It shows how much the price has changed relative to the base.

Formula (individual): Price Relative for item i = (p_{t} / p_{0}) × 100, where p_{t} is price in the current period and p_{0} is price in the base period.

Purpose: Price relatives are used to measure and compare price changes of individual items over time. They are the building blocks for aggregate price indices (like CPI) when combined across items.

Interpretation:

• If price relative = 100 → no change in price (current = base).
• If price relative > 100 → price has risen by (relative − 100) percent.
• If price relative < 100 → price has fallen by (100 − relative) percent.

From individual to aggregate: To obtain an overall price index from price relatives you combine them—either by a simple arithmetic mean (simple average of relatives) or by a weighted mean using appropriate weights (often base-period expenditures or quantities). In official indices, relatives are usually weighted to reflect the importance (expenditure share) of items.

Relation with standard indices: Many index formulas can be interpreted as weighted averages of price relatives. For example, the Laspeyres price index can be written as a weighted average of (p_{t}/p_{0}) where weights are the base-period value shares.

Advantages: Simple to compute for each item; easy to interpret; useful for comparing many items at once. Limitations: Simple averages ignore item importance; different weighting schemes produce different aggregate results.

What are Unweighted Methods?
Unweighted methods of index numbers measure change in prices or quantities between two periods without assigning different importance (weights) to individual items. Every item is treated equally. These methods are simple and useful for quick checks but may be misleading when items have very different importance or quantities.

Common Unweighted Methods

• Simple Aggregative Method
  - Price index (for base year 0 and current year 1): Index = (ΣP1 / ΣP0) × 100, where ΣP1 is the sum of current-year prices of all items and ΣP0 is the sum of base-year prices of the same items.
  - Quantity index: Index = (ΣQ1 / ΣQ0) × 100, where Q denotes quantities.
• Simple Average of Price Relatives (Arithmetic Mean of Relatives)
  - For each item i, compute the price relative = (P1_i / P0_i) × 100.
  - Then take the arithmetic mean: Index = (1/n) Σ[(P1_i / P0_i) × 100] = (100/n) Σ(P1_i / P0_i), where n is the number of items.

When to use: Unweighted methods are used when (a) items are of roughly equal importance, (b) only price lists (not quantities or weights) are available, or (c) a quick, simple measure is needed for classroom exercises or approximate comparisons.

Advantages

• Very simple to compute and understand.
• Requires only price (or quantity) data, not weights.
• Good for preliminary or illustrative analysis.

Limitations

• Ignores relative importance/consumption (weights) of items; can give misleading results if some items are more important than others.
• Sensitive to the choice and number of items included.
• Simple averaging treats large and small price-changes equally even when economically they matter differently.

How to present results and interpret
If index > 100 → prices/quantities have increased on average from base to current year. If index < 100 → they have decreased. Index = 100 → no average change.

Classroom tip: Always name the base year and the type (price or quantity) when reporting the index. Mention which unweighted method you used.

What are weights?
In index numbers, a weight is a number that represents the relative importance (or share) of an item in the total. Weights adjust averages so that items that matter more (for example because people buy more of them or spend more on them) have a larger influence on the overall index.

Why use weights?

• Different items have different economic importance: a big rise in the price of a rarely bought item should not affect the overall price index as much as a moderate rise in the price of a frequently bought item.
• Unweighted averages treat all items equally and can give misleading results. Weights make the index reflect real economic impact (usually expenditure shares).

Common types of weights

• Quantity weights (q): quantities consumed in a reference period (used in Laspeyres when q0 are base-period quantities).
• Value (expenditure) weights (p*q): value of expenditure in a reference period; these are often used to form weighted averages of price relatives.
• Current-period weights (q_t or p_t*q_t): used in Paasche-type indices.
• Chain weights: update weights periodically (e.g., annually) to reflect changing consumption patterns.

Important criteria for selecting weights

• Relevance: weights should reflect relative importance (expenditure shares, quantities) of items to the objective of the index (CPI, WPI, GDP deflator).
• Same period and consistent units: weights must be measured for a consistent reference period and in comparable units.
• Non-negativity: weights should be >= 0; normally positive for items in the basket.
• Stability vs. representativeness: base-period weights (stable) are easy to use but may become outdated; current-period weights are more representative but change over time.

How weights are used (intuition)
Take price relatives r_i = (p_i_t / p_i_0) * 100. A weighted index is an average of these relatives with weights w_i that reflect importance. The simplest general form is:

Index = (sum_i w_i * r_i) / (sum_i w_i)

Choice of w_i leads to different named indices. For example:

• Laspeyres price index uses base-period quantity weights: w_i = q_i0 but, when written as a weighted average of price relatives, the convenient weights are base-period values w_i = p_i0*q_i0.
• Paasche price index uses current-period quantity weights: w_i = q_it (or value weights p_i0*q_it when forming relatives).

Advantages and disadvantages of common weighting choices

• Laspeyres (base weights): Simple, uses historical consumption patterns; tends to overstate cost of living when consumers substitute away from items that become relatively expensive.
• Paasche (current weights): Reflects current consumption and substitution; may understate cost changes compared with a fixed basket because weights change with prices.
• Chain weights: More up-to-date but more complex and may fluctuate.

Worked numerical example

Suppose a two-item basket (Rice and Milk). Base year (0) and current year (t):

Item    p0   q0    pt
Rice    20   10    22
Milk    40    5    50

Base-year expenditure (p0*q0): Rice = 200, Milk = 200, total = 400.

Current-year cost of base basket (p_t * q_0): Rice = 22*10 = 220, Milk = 50*5 = 250, total = 470.

Laspeyres price index = (Σ p_t q_0 / Σ p_0 q_0) * 100 = (470 / 400) * 100 = 117.5. This means prices rose by 17.5% for the base-period basket.

Note: if we form a weighted average of price relatives r_i (in percent) using base-period value weights p0*q0, we get the same result: r_rice = (22/20)*100 = 110, r_milk = (50/40)*100 = 125; weighted average = (200*110 + 200*125) / (200+200) = 117.5.

Summary
Weights are central to making index numbers meaningful. The choice of weights (base vs current vs chain) determines the kind of index and affects whether the index reflects fixed-basket cost changes or changing consumption patterns.

What it is: The Laspeyres Price Index measures how the cost of purchasing a fixed basket of goods and services (the base-year basket) changes over time. It uses base-year quantities as weights, comparing current-period prices with base-period prices.

Core idea: Keep quantities fixed at base-year levels (q0). Compare total expenditure at current prices (p1 × q0) with total expenditure at base prices (p0 × q0).

Formula (summary): P_L = (Σ p1 q0 / Σ p0 q0) × 100. If P_L > 100 → overall price level has risen since base year; if P_L < 100 → it has fallen.

Step-by-step computation:

• Choose a base year and determine the basket of commodities and their base-year quantities q0 and base-year prices p0.
• Collect current-period prices p1 for the same commodities (quantities remain q0).
• Compute total expenditure at current prices using base quantities: Σ p1 q0.
• Compute total expenditure at base prices: Σ p0 q0.
• Compute the index: P_L = (Σ p1 q0 / Σ p0 q0) × 100. The percent change = P_L − 100.

Weighted form: The index can be written as a weighted sum of price relatives: P_L = Σ w0 (p1 / p0) × 100, where w0 = (p0 q0) / Σ p0 q0 are base-year expenditure shares.

Assumptions and properties:

• Uses fixed base-year quantities (no substitution allowed).
• Gives more weight to commodities that had larger shares in base-year expenditure.
• It is easy to compute and stable when current consumption data are poor.
• Downside: tends to overstate cost-of-living increases (upward bias) because it ignores substitution away from goods whose prices rose relative to others.

Comparison (brief): Unlike the Paasche index (which uses current quantities q1), Laspeyres uses base quantities q0. The Fisher Index is the geometric mean of Laspeyres and Paasche and is called an 'ideal' index.

CBSE tip: Remember the formula and the interpretation (use base quantities). For numerical problems, compute Σ p1 q0 and Σ p0 q0 carefully and report the index to two decimal places when required.

Definition: The Paasche Price Index measures the change in the cost of purchasing the current period's basket of goods and services at current prices compared to base-period prices. It uses current-period quantities as weights.

Formula (short): P_P = (Σ p_t q_t / Σ p₀ q_t) × 100, where p₀, p_t are prices in base and current period, and q_t are quantities in the current period.

Interpretation: The index shows how much more (or less) it would cost to buy the current period’s quantities if they were purchased at current prices versus base-period prices. A value >100 indicates an overall rise in prices (inflation) for the current consumption pattern; a value <100 indicates an overall fall.

Key features:

• Weights: uses current-period quantities (q_t).
• Reflects substitution: because weights change with consumption, it can reflect consumers’ substitution toward relatively cheaper goods.
• Tends to give a lower estimate of price rise than Laspeyres in an inflationary environment (because consumers substitute away from goods whose prices rose most).
• Computational difficulty: requires data on current-period quantities for each period, which can be hard to collect for long time series.

Uses in practice: The GDP deflator is an example of a Paasche-type index: Nominal GDP / Real GDP (in base prices) = Σ p_tq_t / Σ p₀q_t.

Advantages and disadvantages (concise):

• Advantages: reflects current consumption patterns, accounts for substitution effects, more representative of present-day spending.
• Disadvantages: data intensive, current quantities may themselves be influenced by current prices (endogeneity), not ideal for fixed-base long-term comparisons.

Computation steps:

1. List goods with their base-period prices (p₀), current-period prices (p_t) and current-period quantities (q_t).
2. Compute numerator Σ p_t q_t and denominator Σ p₀ q_t.
3. Divide numerator by denominator and multiply by 100 to get the index.

Definition: The Fisher Ideal Index is an index number used to measure price or quantity changes between two periods. It is the geometric mean of the Laspeyres and Paasche indices and is considered an 'ideal' index because it satisfies important tests of index-number theory (time reversal and factor reversal).

Intuition: Laspeyres uses base-period quantities and tends to overstate price changes when consumers substitute goods; Paasche uses current-period quantities and tends to understate price changes. The Fisher index takes the geometric mean of the two to reduce these biases.

How to compute (price index): First compute the Laspeyres price index (L) and the Paasche price index (P). Then Fisher price index F = sqrt(L × P). Use the same approach for quantity indices.

Properties:

• Satisfies time-reversal test: the index for 0->1 times the index for 1->0 = 1 (when indices are used in ratio form, not multiplied by 100).
• Satisfies factor-reversal test: Price index × Quantity index = value ratio (Σ p1 q1 / Σ p0 q0), again in ratio form.
• Symmetric and free from systematic bias of L and P.

When it is used: Used in research and some national accounts for chain-weighted measures (e.g., chain-type Fisher index for real GDP) and for international price/volume comparisons. However, statistical offices sometimes prefer Laspeyres for CPI because of data simplicity and historical continuity.

Limitations: Requires price and quantity data for both periods for all items (data-intensive) and can be computationally heavier than simple indices; interpretation requires care when chains of periods are used.

Overview
Other weighted indices are improvements over simple indices (or unweighted indices) because they use quantity (or expenditure) information to give relative importance to items. Two commonly taught weighted indices in Class 11 Economics are the Marshall-Edgeworth index and the Drobisch–Bowley (often written Dorbish–Bowley) index. Both use the arithmetic mean of base-period and current-period quantities as weights to reduce the bias present in Laspeyres (base-weights) and Paasche (current-weights) indices.

1. Marshall-Edgeworth index (ME)
Definition: The Marshall-Edgeworth price index is a weighted aggregate of prices where the weight for each item is the arithmetic mean of the base-period and current-period quantities. It compares the total value of the basket of goods at current prices and base prices, using the average quantity as the weight.

Interpretation: ME tries to balance the extremes of Laspeyres and Paasche by using average quantities. It gives less bias when quantities change between periods and is useful when we can observe both period quantities.

2. Drobisch–Bowley (Dorbish–Bowley) index (DB)
Definition: The Drobisch–Bowley price index is a weighted average of price relatives (price ratios expressed as indices, e.g., (p_t/p_0)*100) where the weights are the arithmetic mean of base and current quantities. In short, DB is the average of price relatives with average-quantity weights.

Interpretation: DB expresses the index as an average of the percentage change in prices (price relatives), each weighted by the average quantity of that item. It is directly comparable to other relatives-based indices and less sensitive to substitution bias than pure base- or current-weighted indices.

When to use these indices

• When quantities are available for both base and current periods.
• When we want an index less biased than Laspeyres or Paasche but do not want to use Fisher's geometric mean.
• For constructing national or sectoral price indices where moderate changes in quantity mix occur.

Key characteristics

• Both use average quantities w_i = (q_0 + q_t)/2 as weights.
• ME uses weighted sums of prices (value approach). DB is a weighted average of price relatives (relative approach).
• They lie between Laspeyres and Paasche values in many practical cases and reduce the bias caused by using only base or current quantities.

Limitations

• Require quantity data for both periods (may not always be available).
• Do not have the exact proportionality and time-reversal properties that Fisher's ideal index has.
• Computation can be more involved than simple indices.

Short worked idea (conceptual): For each item i, compute the average quantity w_i = (q_{0i} + q_{ti})/2. For ME, multiply current prices by w_i, sum and compare with sum of base prices times w_i. For DB, compute price relatives R_i = (p_{ti}/p_{0i})*100 and take their weighted average using w_i.

Overview
Index numbers measure relative change (usually in prices, quantities, or values) over time. Two common methods to present time-series indices are the fixed-base index and the chain index.

Fixed-base Index (also called Base-year Index)
A fixed-base index compares every period to a single, constant base period. The base period is assigned the value 100 and every other period's index shows percentage change from that base.

Unweighted (single item):
I_t = (P_t / P_0) × 100

Weighted (e.g., Laspeyres-type with base-period quantities):
I_t = (Σ p_t q_0 / Σ p_0 q_0) × 100
Here p_t is price in period t, q_0 are quantities in the base period.

Properties

• Easy interpretation: each index tells change relative to one fixed base (e.g., 2012 = 100).
• Good for long-term comparisons when weights (consumption patterns) do not change much.
• Weakness: base-period weights become outdated as tastes, technology, and products change; fixed-base indices can misrepresent true changes over long spans.

Chain Index (also called Chain-weighted or Moving-base Index)
A chain index links (chains) short-period indices (usually consecutive-year indices) to produce an index relative to a reference base. Typically, an index for year t relative to year 0 is obtained by multiplying year-on-year relatives from 1 to t.

General idea (ratios): if r_k = (Value_k / Value_{k-1}), then ChainIndex_t = (r_1 × r_2 × ... × r_t) × 100 = (Value_t / Value_0) × 100. Practically, chain indices are computed using updated weights each link (so each link uses weights of the preceding year), which better reflects changing baskets.

Example of chain linking with weighted link-relatives (chain Laspeyres method):
Link relative for year k (relative to k-1): r_k = (Σ p_k q_{k-1} / Σ p_{k-1} q_{k-1}) × 100. Then the chain index for year t relative to base 0 is the product of the link relatives (divided by 100 for each multiplication except the last), i.e. ChainIndex_t = r_1 × r_2/100 × r_3/100 × ... etc., typically simplified as (Σ p_t q_{t-1} × ... ) representing cumulative change.

Properties

• Weights are updated frequently (each link), so the index better reflects changing consumption/production patterns.
• Reduces bias from obsolete base weights over time.
• Weaknesses: chaining is computationally more involved; chain indices are not additive across long intervals in a simple way and small period-to-period volatility compounds multiplicatively.

Practical guidance

• Use fixed-base when you want a simple comparison to a single reference year (common for published CPI with a chosen base year).
• Use chain indices when relative prices and quantities change rapidly, and you want weights to reflect recent consumption (common in national accounts for real GDP growth).

Worked numerical example (small two-good example)

Fixed-base Laspeyres (base 2018=100):
Denominator Σ p0 q0 = 10×5 + 20×3 = 50 + 60 = 110
2019 numerator Σ p1 q0 = 12×5 + 22×3 = 60 + 66 = 126 ⇒ I_2019 = 126/110 ×100 = 114.55
2020 numerator Σ p2 q0 = 15×5 + 25×3 = 75 + 75 = 150 ⇒ I_2020 = 150/110 ×100 = 136.36

Chain Laspeyres (update weights each year):
Link 2019 (relative 2018→2019) using q0: r_2019 = 126/110 ×100 = 114.55
Link 2020 (relative 2019→2020) using q1: Σ p2 q1 = 15×6 + 25×2 = 90 + 50 = 140; Σ p1 q1 = 12×6 + 22×2 = 72 + 44 = 116 ⇒ r_2020 = 140/116 ×100 = 120.69
Chain index 2020 relative to 2018 = r_2019 × (r_2020/100) = 114.55 × 1.2069 ≈ 138.36

Note the difference: fixed-base I_2020 = 136.36 while chain I_2020 = 138.36. The chain index is higher here because the updated weights (q1) put relatively more weight on the good that rose more in price.

Summary
Fixed-base: simple, single benchmark, but may become outdated. Chain: updates weights frequently, more responsive to change, but requires more computation and interpretation.

What are index numbers of quantity and value?

Index numbers are simplified measures that show relative changes in a variable (price, quantity or value) over time with a chosen base period = 100. Index numbers of quantity measure how physical amounts (units produced, sold or consumed) change over time; index numbers of value measure changes in the money value (price × quantity) of those items.

Why they matter: Quantity indices help identify real changes in output or consumption (removing price effects). Value indices measure changes in total receipts, sales or GDP in nominal terms. Together they help separate whether total value changes are due to price changes, quantity changes, or both.

Basic construction ideas

• Single-item index (for commodity i):
• Quantity index: Qi = (q1 / q0) × 100
• Value index: Vi = (p1·q1 / p0·q0) × 100
• Aggregate (many commodities) — two common approaches:
• Simple aggregate quantity index: Q_simple = (Σ q1 / Σ q0) × 100 (all items equally weighted by quantity)
• Aggregate value index: V = (Σ p1·q1 / Σ p0·q0) × 100 (direct comparison of total value)

Weighted quantity indices (use prices as weights to reflect importance):

• Laspeyres quantity index (base-period weights): Q_L = (Σ p0·q1 / Σ p0·q0) × 100. This asks: how much would base-period-priced value change if only quantities changed?
• Paasche quantity index (current-period weights): Q_P = (Σ p1·q1 / Σ p1·q0) × 100. This asks: how much would current-period-priced value change if only quantities changed?

Relationship between price, quantity and value

• For a single commodity: Value change = Price change × Quantity change. Numerically, (p1·q1)/(p0·q0) = (p1/p0) × (q1/q0). So the single-item value index equals the product of its price and quantity indices.
• For an aggregate of many commodities this simple multiplicative decomposition does not generally hold because of differing weights. Aggregate value change is

V = (Σ p1·q1) / (Σ p0·q0)

But this is not simply (aggregate price index) × (aggregate quantity index) unless specific weighting conditions hold. That is why weighted indices (Laspeyres, Paasche, Fisher) and careful interpretation are necessary.

Practical notes

• Base period: choose a convenient normal year; index values are usually reported with base = 100.
• Interpretation: index > 100 indicates increase from base; < 100 indicates decrease.
• Use cases: quantity indices for industrial output, crop production, or physical volume of sales; value indices for nominal GDP, export/import values, and total revenue.
• Limitations: choice of base and weights affects results; simple aggregates ignore relative importance (prices or expenditure shares).

How to compute (step-by-step for aggregates)

1. Collect p0, q0 for base period and p1, q1 for current period for each commodity.
2. Compute unit values v0 = p0·q0 and v1 = p1·q1. Then aggregate: Σq0, Σq1, Σv0, Σv1.
3. Simple aggregate quantity index = (Σq1 / Σq0) × 100.
4. Aggregate (value) index = (Σv1 / Σv0) × 100.
5. If a weighted quantity measure is required, compute Laspeyres or Paasche as given above.

Interpretation example (summary): If aggregate value index = 108 and quantity index = 106.7, the remaining part of value increase is due to price rise (but decomposition must use consistent weighting to precisely split price and quantity effects).

Definition: The Cost of Living Index Number or Consumer Price Index (CPI) is a measure that shows the change in the cost of maintaining a given standard (fixed basket) of living over time. It expresses the ratio of the cost of the same basket of goods and services in the current period to its cost in a chosen base period, usually scaled so that the base period index = 100.

Purpose and Uses:

• Measure inflation (rate of change in prices faced by consumers).
• Adjust wages, pensions, rent agreements and tax brackets for changes in purchasing power.
• Inform monetary and fiscal policy decisions.

Basic Idea / Construction Steps:

1. Choose a base period and fix the basket of goods and services that represents typical consumer consumption (quantities q0).
2. Collect prices of each item in the base period (p0) and in the comparison/current period (pt).
3. Compute the total cost of the basket in base period: Σ p0 q0 and in current period: Σ pt q0 (Laspeyres approach).
4. Form the index: CPI = (Cost in current period / Cost in base period) × 100.

Common Formula Types:

• Laspeyres CPI (fixed-quantity): uses base-period quantities as weights — widely used for CPI construction.
• Paasche index: uses current-period quantities (less common for CPI).
• Weighted price relatives: CPI can also be written as a weighted average of price relatives, where weights are expenditure shares (w_i) from the base period.

Properties and Limitations:

• Reflects price changes for a fixed consumption pattern; does not capture substitution by consumers when relative prices change (substitution bias).
• May become unrepresentative over time unless the basket and weights are updated periodically.
• Quality changes and new goods require adjustments (quality bias and new-good bias).

Practical Notes: Statistical agencies typically publish several CPIs (e.g., urban, rural, combined) and periodically revise base year and weights so the CPI stays representative.

What are Index Numbers? Index numbers are statistical measures that show changes in a variable (usually prices or quantities) over time relative to a selected base period. "Steps in Construction" describes the systematic procedure used to build a reliable index number.

1. Define the purpose and scope. Decide whether you need a price index, quantity index or value/aggregate index, and whether it will measure inflation, cost-of-living, production, etc. The purpose determines selection of items, weights and formula.
2. Choose the base period. Select a normal (typical) year as base. All comparisons will be made relative to this period (index = 100 in base year by convention).
3. Select the items (basket). Identify the representative goods and services that make up the basket (e.g., food, rent, transport for CPI). Items must reflect typical consumption/production relevant to the purpose.
4. Decide weights and their source. Determine importance (weights) for each item—usually based on expenditure/quantity shares in the base period or an average period. Weights ensure items that matter more have bigger influence.
5. Collect data (prices and/or quantities). Gather reliable price and quantity data for base and current periods from surveys, stores, markets or official sources. Ensure consistent item definitions and quality adjustments.
6. Choose the formula/method. Decide whether to use a simple (unweighted) method or a weighted formula: common weighted formulas are Laspeyres, Paasche and Fisher (Ideal). Choice depends on data availability and properties required (e.g., satisfying certain reversal tests).
7. Compute intermediate values (price relatives, totals). Calculate price relatives (current price/base price × 100) for items and multiply by weights if weighted method is used. For aggregative methods compute sums (sum of weighted prices, sums of expenditures).
8. Calculate the index. Apply the chosen formula to get the overall index number. Convert into index form (base = 100) and, if necessary, chain-link indices across different base years.
9. Check and test the index. Verify calculations, check for outliers, and apply consistency tests (e.g., time-reversal and factor-reversal properties) where appropriate. Adjust for quality changes if needed.
10. Interpret and report results. Present index values, percentage change (inflation rate), and decomposition (which items contributed most). Document methodology, base year, and limitations.

Key cautions: choose representative items and accurate weights, maintain consistent item definitions, adjust for quality changes and seasonal effects, and state limitations when interpreting changes as cost-of-living or welfare changes.

Overview: Tests (also called axioms or desirable properties) are logical checks that an index number formula should satisfy to be considered good. They show whether an index behaves sensibly under certain changes in prices and quantities. The main tests studied at Class 11 level are: Identity test, Proportionality test, Time‑reversal test and Factor‑reversal test. (Two additional ideas sometimes mentioned: Circular/Transitivity test and Commensurability test.)

1. Identity Test

• Statement: If prices in the current period (period 1) are exactly equal to prices in the base period (period 0) for every commodity, the price index should be 1 (or 100 when the index is expressed on base 100 scale).
• Why it matters: If nothing has changed in prices, the index must show no change.
• Mathematical form: If p1i = p0i for all i, then I01 = 1 (or 100).

2. Proportionality (Scalar) Test

• Statement: If all prices in period 1 are changed by the same proportion k (p1i = k·p0i for all i), the index should equal k (or 100·k).
• Why it matters: A uniform proportional change in all prices should be reflected exactly by the index.
• Mathematical form: If p1i = k·p0i for all i, then I01 = k (or 100·k).

3. Time‑reversal Test

• Statement: If we reverse the roles of base and current period, the two indices should be reciprocals. In ratio form: I01 · I10 = 1. In base‑100 form: I01 · I10 = 10000.
• Why it matters: The measure of change from 0 → 1 should be the inverse of change from 1 → 0.
• Which indices satisfy it: The Fisher ideal index satisfies time‑reversal (exactly). Laspeyres and Paasche do not satisfy it in general.

4. Factor‑reversal Test (Product Test)

• Statement: Price index times corresponding quantity index should equal the value (expenditure) relative. In symbols: P01 · Q01 = V01, where V01 = (Σ p1 q1)/(Σ p0 q0).
• Why it matters: Total value change can be decomposed into a pure price effect and a pure quantity effect; their product should give the total value change.
• Which indices satisfy it: The Fisher index (price and quantity forms) satisfies factor‑reversal. Also Laspeyres price index times Paasche quantity index equals the value relative (and vice versa).

Remarks on other tests:

• Circular (or Transitivity) Test: For three periods A, B, C, the product I_AB · I_BC · I_CA should equal 1. This is a strong property and usually not satisfied by simple indices.
• Commensurability Test: Index numbers should compare like with like (units, quality); practically it means index should be based on comparable items and correct weights.

Common index formulas (brief):

• Laspeyres price index (L): L01 = (Σ p1 q0) / (Σ p0 q0).
• Paasche price index (P): P01 = (Σ p1 q1) / (Σ p0 q1).
• Fisher ideal price index (F): F01 = sqrt(L01 · P01) (geometric mean of L and P).
• Value relative (V): V01 = (Σ p1 q1) / (Σ p0 q0).

All the above tests are useful when choosing or evaluating an index for practical use (CPI, GDP deflator etc.). The Fisher ideal index is called 'ideal' because it satisfies both time‑reversal and factor‑reversal tests, and also the identity and proportionality tests.

Summary table (which tests are satisfied by common indices):

• Identity: Laspeyres, Paasche, Fisher — all satisfy.
• Proportionality: Laspeyres, Paasche, Fisher — all satisfy.
• Time‑reversal: Only Fisher satisfies in general.
• Factor‑reversal: Fisher satisfies; Laspeyres·(Paasche quantity) = Value relative (so a pair can satisfy).

Overview: Index numbers (price index, quantity index, CPI, WPI) summarise relative changes in a group of variables over time. While useful, they have several problems and limitations that affect accuracy, interpretation and policy use.

• Choice of base period: The base year chosen affects index levels and growth rates. If the base year is atypical (price shocks, supply disruptions), the index can be misleading. Adjustment: periodic rebasing or chain indices.
• Selection and representativeness of items: The basket of goods/services used to construct an index may not represent all consumers (regional, income groups, changing consumption patterns). This leads to bias when applying a single index nationally or across groups.
• Weighting problems: Aggregated indices require weights (expenditure shares or quantities). Wrong or outdated weights distort the index. Different weighting methods (Laspeyres, Paasche) produce different results.
• Substitution bias: Fixed-weight indices (e.g., Laspeyres) ignore that consumers substitute cheaper goods for expensive ones, causing overestimation of cost-of-living increases. Adjustment: chain indices or Fisher index reduce this bias.
• Quality change and new goods: Improvements in quality (e.g., smartphones, cars) make direct price comparisons invalid — higher prices may reflect higher quality. New products enter the market unaccounted for in the original basket. Hedonic adjustments or quality-adjusted prices are necessary but complex and sometimes subjective.
• Outlet and measurement changes: Changes in where goods are bought (online vs. retail), measurement errors, recording mistakes, or different units (kg vs. packet) produce inaccuracies.
• Seasonal effects and timing: Seasonal items (fruits, vegetables, holidays) fluctuate widely. If not seasonally adjusted, short-term indices can be misleading.
• Aggregation paradoxes and formula choice: Different index formulas (simple average, Laspeyres, Paasche, Fisher) give different values; there is no single "true" index. Aggregation across heterogeneous products may hide divergent movements within groups.
• Interpretation limits: Index numbers show relative change (percent change) not absolute levels. A price index rising by 50% does not tell which goods or income groups are most affected. It also cannot alone measure welfare or distributional impacts.
• Data collection costs and timeliness: Producing accurate indices requires frequent, high-quality price and quantity data. This is costly and often causes delays or sampling compromises.

Practical consequences: Policymakers relying on indices for inflation targetting, wage adjustments or social benefits may misjudge real living costs if these problems are ignored. Users must know formula choice, basket composition, rebasing frequency and any quality adjustments used.

What this topic covers
Applications and interpretation of index numbers show how index measures (like CPI, WPI, GDP deflator) are used to summarize relative changes in prices, quantities or values over time or across places and how to read and draw conclusions from them.

Main applications

• Measuring inflation: Consumer Price Index (CPI) and Wholesale Price Index (WPI) track changes in price level. The percentage change in the index gives inflation rate.
• Cost of living adjustments (COLA) and wage indexation: Salaries, pensions and tax brackets can be adjusted using CPI to keep real purchasing power constant.
• Converting nominal values into real values: Remove price effects from money values (wages, GDP, revenue) to compare volumes across time.
• Policy and business decisions: Central banks, governments and firms use index numbers to set interest rate policy, update budgets, price contracts and plan production.
• Comparative studies: Compare price levels across regions or time (cross–section and time–series comparisons) and test relative performance of sectors.
• Deflators and national accounts: GDP deflator transforms nominal GDP into real GDP to measure real growth.

How to interpret movements

• Level of index: If index = 100 in base year, index = 120 means prices are 20% higher than base year.
• Percentage change between two periods: Tells inflation or deflation rate over that interval.
• Short-term spikes vs trend: A single high monthly change may be seasonal or transitory; look at year-on-year and moving averages to identify trend.
• Base year effects: Changing base year re-scales index levels but percentage changes remain comparable if re-basing is done correctly.
• Weighted vs simple index: Weighted indices (like CPI) reflect relative importance of items; interpreting changes requires understanding weights.
• Limitations to keep in mind: substitution bias, quality changes, introduction of new goods, changes in consumption patterns and choice of base year can affect interpretation.

Practical reading tips

• Prefer year-on-year (12-month) percent change to avoid seasonality for monthly data.
• Look at both nominal and real series (e.g., nominal wages and real wages) to judge welfare changes.
• When comparing sectors or regions, ensure indices use comparable baskets and base years or re-base/chain them.

Simple interpretive example (explained):
If CPI = 180 in 2025 and CPI(base year 2015)=100, then general price level is 80% higher in 2025 than in 2015. If nominal average salary rose from 50,000 to 70,000 over same period, real salary change = (70,000 * 100 / 180) - (50,000 * 100 / 100) expressed relative to base or compute percent change in real terms.

What this topic covers
Practical Computation and Numerical Exercises teaches how to construct and interpret index numbers (price and quantity indices) using concrete numerical procedures. It focuses on stepwise computation, choice of base year, use of weights, and comparison of different formulae (simple aggregative, weighted/Laspayres, Paasche, Fisher, CPI construction and chain linking).

Key steps in practical computation

1. Select the base year and current (comparison) year.
2. Decide the items (basket) and collect prices (and quantities or expenditure weights).
3. Compute price relatives for each item: (P1/P0)×100.
4. Choose a method to aggregate relatives: simple average, simple aggregative, or weighted (use quantities or expenditure shares as weights).
5. Compute the index (expressed with base = 100). Convert index change to percent change: (Index − 100)%.
6. When required, compute quantity indices similarly (swap roles of price and quantity) and check consistency (e.g., Fisher ideal index).

Common methods (when to use)

• Simple aggregative index: quick rough comparison when no weights are available.
• Weighted (Laspeyres) index: common for price indices (uses base-year quantities as weights).
• Paasche index: uses current-year quantities; useful when current consumption pattern is relevant.
• Fisher ideal index: geometric mean of Laspeyres and Paasche; often preferred for theoretical properties.
• CPI (Consumer Price Index): uses a fixed basket and expenditure weights to measure cost of living changes.

Practical tips

• Always state the base year and basket clearly.
• Use consistent units and rounds only at the final step to avoid cumulative rounding error.
• When given expenditures (value = price × quantity), you can derive weights directly from base-year expenditures for Laspeyres.
• For time-series exercises, chain indices (link relatives year-to-year) if baskets or bases change over time.

Common pitfalls
Confusing price relatives (item-level) with aggregate index; using current-year quantities for Laspeyres (or vice versa); forgetting to multiply by 100 to express indices; inconsistent weights (must sum to 1 or be used consistently in numerator and denominator).