Significance

General level of prices for goods and services in an economy. If all prices stay fixed for a while, the price level is unchanged, too. When inflation takes place, the price level arises.

Computation

If in an economy were only one good, the price level would be directly the price of that good. Already with two goods, the problem begins to be not so easy.

If both goods see their prices increased by the same percentage amount, the price level can be said to be increased by that growth. But if dynamics differ, the changes in general price level should be extracted from those divergent paths. In particular, if one price soared and the other fell, what happened to the price level? Was it untouched?

More than one answer can be given and conventions only can fix the problem but not solve it. Indeed, different people could continue to disagree and prefer certain conventions over others.

Let's see an example:

Passing from period 1 to 2, the price of X has increased, while Y price has fallen. In the first period, Y has been bought more than X (2.5 times more, to be exact) and the more so in the second period (6 times more).

What is the price level at period 1? Well, there are at least 4 possible answers. How much did the the price level change? Not less than 5 possible outcomes!

To follow the discussion on how to calculate the general price level and to introduce your own changes, you can open this MS Excel file.

Let's shortly see some possible definitions.

Price level n. 1: the simple arithmetic average of all current prices
Price level n. 2: the weighted arithmetic average of current prices, with quantities in period 1 as weights
Price level n. 3: the weighted arithmetic average of current prices, with quantities in period 2 as weights
Price level n. 4: the weighted arithmetic average of current prices, with quantities in the current period as weights (variable weights with time elapsing).

Price level changes: changes in the 4 abovementioned levels plus a direct simple averaging of all prices dynamics.

The resulting dynamics are the following:

As you can see, the same situation can be described quite differently by various computations of price level, where quantities and prices are both involved.

With more than two goods, things even get more complicated, not to mention what happens when one should consider the evolution of the quality level over time, as you can see in this free software.

The convention that central statistical offices usually adopt with many goods is the number 2, although international authorities sometimes would even support further solutions (as the so called "Fisher's formula").

As a consequence, the overall price level is not directly observed by economic agents (consumers, firms,…). It is computed by statistical offices according to conventions and the economic agents will rely on announcements, up to the extent they trust them.

Economic agents pay much attention to the prices of the goods they actually buy or sell. Different people with widely different consumption structures may face different price level and dynamics (more or less inflation).

Different industries will decide their prices autonomously, with some "inflation leaders" and other "inflation losers".

Whereas everybody know the same nominal prices, inflation-corrected "real" values will always depend on conventions, where different agents may have divergent interests in adopting one or another.

In particular, when wages negotiations try to connect the changes in wages to inflation, business will prefer conventions that reduce the computed inflation, while trade unions the opposite.

Composition

The price level is a certain average of the prices of a broadly defined group of commodities.

Depending on the chosen commodities and the considered step of the distribution channel, one gets the following, among the others:

1. consumer price level;
2. wholesales price level;
3. producer price level.

Another technique to get price level indicators are the "deflators". Computed as a ratio between current and constant-price values of the same aggregate (as you can see in the MS Excel example), one usually consider:

4. GDP deflator;
5. investment deflator;
6. consumption deflator;
7. public expenditure deflator.

In any case, wages and income sources are never included in the basket that real statistical offices adopt. This is a far cry from neoclassical microeconomic description of inflation as a contemporary proportional increase of all prices and incomes [1].

Similarly, assets as shares and real estate are not included in the common price level definitions. Thus, an increase in stock exchange index is usually not defined as "inflation".

Determinants

Prices and quantities of all goods and services determine the overall price level. Whatever the convention adopted, an increase of price level is due to an increase of at least some prices.

Prices are chosen by sellers, with the exception of specific markets dominated by buyers and/or requiring negotiations (often for both price and product features).

Sellers, usually firms, take into account mainly the following elements:

1. costs,
2. desired margins,
3. competitors' prices,
4. quality comparison with competitors' products;
5. consumer propensity to pay,
6. consumer response in general;
7. expectations about future events;
8. their urgency to sell;
9. the further perspective sales with the same client and with clients positively impressed by seen the product adopted by the first client;
10. the phase in the product life cycle.

For instance, penetration prices are particularly low level of prices to attract early customers. Cream-skimming prices are particularly high prices to position the product in the top-level segment of quality and to attract top-level consumers, possibly with the perspective of reducing the price to get more clients later on.

The general price level depends first and foremost by individual decisions of firms. For most common consumer goods, prices are determined by producers and by distribution channels (together or separately). Unconstrained monopolists are often particularly aggressive in raising prices. If a good is without any close substitute is likely to have relatively a high price. The composition of price basket mediates the effect of individual prices on the overall price level. If some prices fall and other raise, the average price level will fairly represent a relatively small fraction of actual price changes. By contrast if most price go in the same direction by the same percentage amount, the average price level will be highly representative of actual changes.

The variables having an impacts on the elements taken into account by decision-makers are an indirect second level of determinants of price level. Cost pressures, coming for example from wages or from imports after a devaluation, can exert a powerful influence on it, although it will be possibly mediated through productivity considerations.

Income distribution and the decision-making rules of consumers play an important role in determining demand curves before which an informed seller could set profitable prices. If this happens on many markets, the overall price level can be influenced by the levels in income and consumption expenditure, as you can test using these data.

Impact on other variables

Inflation is the main variable that depends on price level. An increase of price level without a corresponding wage increase will reduce the real income of workers.

Real money will be reduced by a price level increase, provided nominal money does not grow by the same amount.

Since an increase of price level depends on changes in single prices, it can be accompanied by changes in the price-sensitive structure of consumption, investment, public expenditure, and so on. At the same time, these aggregates are dramatically different in structure, thus their specific deflators are usually strikingly different for each of them. For example, public expenditure may face higher inflation than other aggregates.

The decision to save may be influenced by price level expectations.

To the extent exporters charge the same price to foreigners, a higher price level would make their products more expensive and less competitive. However, there is some crucial piece of evidence that domestic and export prices are different, in particular that "exporters systematically charge higher prices for comparable goods when exporting to high-income countries. On average the richest country in our data set pays 40 percent more for the same good than the poorest country. We perform robustness checks that support our interpretation of this as true price discrimination, and not the result of quality differences or transfer pricing of related party trade".

Long-term trends

In most countries in the XXth century, price levels have always grow following an exponential path. A few countries in deep recession have manifested a falling general price level. In previous centuries, long stagnation and falling price levels have often compensated periods of rapid price growth, resulting in a contrasted dynamics.

Business cycle behaviour

Price level usually grows all the time during the business cycle. If wages are frozen, large increases in certain prices produce a crisis in the other sectors, possibly leading to a recession. In some deep and prolonged recessions, the overall price level may fall.

Data

Income, consumer expenditure, and price level in a regional comparison

Inflation rates for 170 countries (1970-1996)

93 Food products prices in 198 countries (1985-2001)

Cocoa beans world prices (1971 - 2001)

Coffee world monthly prices (1982 - 2000)

Price comparision across categories and specific items

Formal models

Price levels and inflation in a 2-goods closed economy

An interactive map of how the economy works according to a basic macroeconomic scheme: the IS-LM model

Price determination in monopoly

Price level in dynamic competition with rising quality levels in good features and with bounded rational consumer agents

Price level changes and their impact on consumer's utility in the neoclassical model

NOTES

[1] Indeed, to understand the so-called "neutrality of money", try this easy experiment with our free software: increase by the same percentage both prices of X good and Y good as well as income. Nothing will change in the budget line, thus in the chosen bundle. If, however, income does not increase, a painful fall of utility will take place.