Skip to content

GDP: Income Side (11:00 am)

Here is a table of the income side of our GDP accounting, with data for 2017Q3, the most recent available. You can find much, much greater detail on the NIPA portion of the Bureau of Economic Analysis web site. What you can see is that compensation is the largest single component. Of course we might think that should be combined with small business income of $1,382 billion. Other changes? Observations?

Gross domestic income 19,515 100%
compensation 10350 53%
wages & salaries 8388 81%
supplements (social security contributions) 1962 19%
excise & import taxes less subsidies 1270 7%
net operating surplus ("profits") 4842 25%
net interest 805 17%
transfer payments (social security taxes) 153 3%
small business income* 1382 29%
rental income 747 15%
corporate profits less government business 1755 36%
taxes 476 27%
dividends 764 44%
retained earnings 527 30%
depreciation (“consumption of fixed capital”) 3052 16%

Of course we would like to know how things changed over time. I could for example look up the change in compensation of workers as a percent of GDP. For some questions, that is the proper metric. But we really want to know how we're doing, how compensation has changed over time. The graph I add uses "real" data, that is, corrected for price changes. We'll get to the nominal / real distinction on Friday.

Note that in the following graph I use a log scale. That's preferable when there's a growth trends and the magnitude changes a lot. If X is the level, then a 10% increase is X*(1+.10). If we take a log then that becomes log X + log 1.1. So the same vertical increment represents the same percentage increase, whether the level of pay is low or high. If we use a regular scale, then a modest vertical increment to pay in 1960s represents a large percentage increase, while the same vertical increment in 2017 represents a small one. In other words, we can't "eyeball" what is going on. Another takeaway: with a log scale, the slope of the line is the (percentage) rate of increase. Again, your eyes fail you with a regular scale, as the slope doesn't translate directly into the rate of change.

Short-cuts: each item, as with any accounting system, incorporates lots of detail about what is and isn’t measured. Here is one example. Small businesses that aren’t set up as formal corporations don’t pay their owners a salary. Instead, earnings are simply part of personal income, and included as one part of overall income on Federal Income Tax Form 1040 (Schedule F for farmers, Schedule C for other businesses). A hint as to the technical accounting aspects: the formal line item is “Proprietors’ income with inventory valuation and capital consumption adjustments.”

16 thoughts on “GDP: Income Side (11:00 am)

  1. ingramk20

    The graph representing "Real Compensation per Hour" over time really clearly demonstrates the overall upward trend while also allowing the user to scroll and see what quarters of each year workers experienced decreased hourly compensation. I like the interactive features of FRED that help me understand the data even better.

    1. ingramk20

      UPDATE: I thought we were just supposed to write a comment to verify everything was working. Here is my actual comment

      The slope of the line from the early 2000s to the present suggests wages have stagnated (or at least risen at a much slower pace) in the past 20ish years. From around the 1960s to the 80s wages rose continuously at a consistent pace. The line also gets less smooth as we move on in time and it appears wages have become more volatile from quarter to quarter. Does this have anything to do with the shift of the makeup of jobs/ labor? While the overall trend is still a rise in wages (even if an extremely modest one), I'm incredibly surprised at the volatility in recent years and wonder whether this trend will continue in the future.

      1. minsong20

        On that note, it is interesting that the changes of the 90s brought wages down, but the recession left things stagnant.

  2. Molly Mann

    I think this interaction between accounting and economics through income is fascinating. It amazed me to see the large percentage social security made up of wages paid out. I’m also curious to see the impact of real vs nominal on compensation per hour.

  3. pinhof18

    Percentage of depreciation in the U.S. seems pretty significant. How does that compare to other Western countries?

  4. leebernstein

    I think the fact that social security (transfer payments) only make up 3% to be incredibly interesting. It is typically discussed that social security is where a large amount of people's money is going, but overall this is showing that thats not the case.

  5. Mariam Samuel

    I'm wondering if the percent of compensation from supplements should be a bit higher if 20% of people are receiving social security currently. I would also think contributions should reflect the trend of increasing compensation shown in the graph.

    I find it interesting to compare the quarterly changes in Real Compensation per hour to the overall trend in the graph. This comparison stresses the importance of long term data, especially in economies which experience drastic changes from month to month.

    To connect the graph and table, I would predict the GDI in 2019 significantly (statistically) larger than $19515 billion. Based on the quarterly trends in the graph, I would also predict a slight increase in the compensation portion of the CDI for Quarter 4.

  6. dodsonm20

    The graph displaying change in compensation of workers as a percent of GDP does show a decreasing positive slope but I am curious to see another visualization that displays the national value for minimum wage over the same time period. It'd be interesting to see how the influence of government has helped or hurt the change in compensation.

  7. skinnerf20

    The idea proposed about moving small business income into compensation is an interesting question. This change would bring compensation's percentage of total GDI closer to the standard 2/3rds value it typically has.

  8. bullr20

    As the graph indicates, it is apparent that wages have increased steadily over the last 80 years. This data is "real" and therefore it reflects an actual increase GDP per capita, adjusted for external price changes (i.e. inflation). It is startling to recognize (or see represented graphically) that economic growth literally translates to each generation having more income than their predecessors, a fact so intuitive it often flies under the radar.

  9. Edward Calley

    Interesting that the rate of increase for compensation has seemed to level off in the past 15 years or so. It would be interesting to see how the make up of gross domestic income has changed over the years.

  10. dugganj20

    It is amazing to see the increase of worker compensation throughout the decades. However, one must remember that as compensation increases, so does the expenses of living such as food and housing. Thus, the increase of compensation over time does not necessarily mean people are getting richer, it is a reflection of how living expenses increase at the same time.

  11. clintong20

    I find amazing how the worlds between politics and economics collide within the world of macroeconomics. When analyzing this data, I could not help but to think about how laws or mores have a direct impact on the business cycle. I believe the price of living is increasing, not the compensation of Income.

  12. calihanj20

    I find it very important that real wages were used in the graph. Since inflation negates some of the gross wage increases, real wages must be used to demonstrate purchasing power more closely. While slowing wage growth is not ideal, we can see that the US worker has more purchasing power now than ever before.

  13. riversc20

    The log scale of the second graph is a powerful tool when looking at data; I find it easier to see what is going on long-term. As you mention, the slope translates more directly into the rate of change when data is shown in a log scale versus a regular scale.


Leave a Reply