додому Finance & Business Economy How the wholesale price index is actually formed

How the wholesale price index is actually formed

Numbers don’t lie, but they can obscure reality. See historical data for the United States, United Kingdom, Germany, and France from 1790 to 1940. A jagged wholesale price index line will appear. This is a measure of relative price change. A series of numbers representing the average change in expenses between time periods. Or across places.

These tools were born out of necessity. Early developers need to know the cost of living. It is necessary to calculate how much wages need to be raised so that employees are not left behind. This is still a core feature. Use these indices to evaluate price changes over time. Use these to compare costs between different countries. The consumer price index is its best-known cousin, but the wholesale index tells a different story one of production, not just consumption.

Sampling problem

Data collection is the biggest problem. It is not possible to track all prices around the world. It’s not possible. Sampling is necessary in most cases. The larger the country, the more complex the sampling pattern.

Imagine trying to build an index for a large country. I need a sample that represents price changes in different cities. I need information about different types of sales points. supermarket. Department store. A nearby store. And also from different products. Ideally, the number of prices selected for each category would correspond to its importance in the country’s total spending. Most indices are only approximations to this ideal. They are close enough to be useful.

Consistency trap

When you get the sample, you need to collect prices. The goal is simple. The difference between the two dates should only reflect the price change. Nothing else. Ideally, you would collect the exact same items each time.

Some products are easier to standardize. Let’s take wheat as an example. No. 2 Red winter. Bulk. Carlots. FOB Chicago. spot market price. Average of high and low prices. per bushel. If all products were this accurate, indexing would be easy. However, most of the products are not wheat.

It is difficult to describe the product well enough for different agents to find the identical item. “Men’s shoes” is not a specific object. This is a class. Therefore, the price collection agencies are rely on respondents. Usually a commercial enterprise. Reports the price of the same variation in consecutive periods. The variant might differ from company to company, but as long as the companies follow the same version of the product, the data holds up.

Quality drifts. A 1920 shoe is not the same as a 1930 shoe, even if the label says the same. To overcome this problem, the agency avoids calculating the average observed price for each product by date. Instead, convert each price to a percentage of the previous period’s price from the same source. These are called price relatives. They isolate change. Eliminate the noise associated with product definition changes.

The weight dilemma

Now you have relatives. The next step is to combine them. Average relatives of the same specification from different suppliers. In some cases, we first calculate the city-specific average and then combine the cities. This also works for similar items.

The real difficulty is combining relatives for different commodities. You have to give them different weights. Not all products are equal. In the wholesale index, wheat prices are more important than pepper prices. Wheat fed people. Pepper flavors them. The economic weight is different.

But here’s the problem. The relative importance of products changes over time. Some items are no longer in use. New ones appear. The composition and design of the items are so different that you wonder if they are the same item. Under these conditions, the weights selected are only accurate for one period.

Which period? Base year.

The further you move from the base period, the less meaningful the comparison becomes. You are comparing the modern economy to the weighted baskets of decades ago. These weights become historical artifact. Price indices can only measure price changes relatively accurately for periods close together in time.

This creates blind spots. If you look at it over a 100-year period, as Burns and Mitchell do, you see a renaissance. Best guess. Indexes are valuable. They show trends. But they also hide structural changes beneath the surface. The basket changes. The weights rot. The numbers remain, but their meaning changes.

How much of the inflation you see in these historical charts is real and how much is just a statistical artifact trying to measure a moving target? Usually the answer is both. The indexes tell us where prices went. It is difficult for them to tell us why the basket itself no longer looks like an economy.

Invisible taxes due to quality changes

The real problem with price indexes isn’t just math. The world does not stand still.

In an industrialized economy, grandpa’s toaster is not your toaster. Physical and technical characteristics have changed so dramatically that true identity has become a myth. However, price indices attempt to track inflation through “like-for-like comparisons”. As the apple is constantly evolving, this metric will break.

There is no perfect solution for quality changes. You cannot remove products from your shopping cart. systematically remove goods that have improved pushes the index upwards. Improvements are good. Removing them would make inflation appear higher than it actually is. If you remove deteriorating goods, the bias flips. The index looks better than it actually is.

So statisticians are trying to make changes. They look at cost data. For example, car manufacturers can estimate how much it will cost to add new safety features or improve fuel efficiency between years. Cost increases are considered quality changes. The remaining price difference is treated as pure inflation.

However, this approach has its drawbacks. It misses value that doesn’t cost extra.

Improvements that improve user experience without increasing production costs are invisible to traditional adjustment models.

This caused a debate. Has the US government been exaggerating inflation for decades? In 1961, the Stigler Commission concluded that most economists believed there was a systematic upward bias. The logic is simple. goods get better. If you don’t take this into account, you will have to pay for better quality as much as for raising prices. This bias may also exist in other countries as the US index is often considered the gold standard.

The Bureau of Labor Statistics disagrees. Their official position is that errors due to quality changes cancel each other out. Some products are being improved. Some get worse. The net effect is zero. It’s a valid position, but it feels more like a hedge against criticism than a rigorous proof.

List price and actual price

The second source of error is simpler. Many indices rely on list prices. This is the sticker price. Recommended price. These are not the amounts that people actually pay.

List prices change slowly. They are sticky. However, the actual purchase price varies from moment to moment. They fluctuate. They drop.

An examination of public procurement departments reveals this deficiency. The buyer paid less than the stated list price. The variations are even greater. The offer price is a basis for negotiation and is not the result of the transaction.

When indices use list prices, they miss the discounts. They miss the sales. They miss the reality of the market. This causes distortion. Inflation looks higher than the cash leaving your pocket.

Which is more accurate? Printed label or receipt? The answer determines whether you feel the pressure of inflation or whether you just read it in the report.

Exit mobile version