Subject: Dallas Plant Productivity

MEMORANDUM[mwf1]

Date: [mwf2]September 2, 2001(revised 1/10/05)

Subject: Dallas Plant Productivity

To: Carl Sawyer[mwf3]From: Matt Ford

______

Introduction[mwf4]

As requested[mwf5], productivity of the Dallas plant has been investigated. The focus of the investigation was on productivity levels and trends from 1999 to 2000. Findings are compared to national productivity data[mwf6].

Findings[mwf7]

Productivity results. [mwf8]Partial factor productivity increased about 4% [mwf9]from 1999 to 2000. Large increases in labor and resin productivity were offset in part by a decrease in capital productivity.

National comparison. The Dallas labor factor productivity increase of 9% exceeds current national average productivity increases of between 2-3%.

Tracking system. Develop[mwf10] a system to measure productivity changes on a quarterly basis. The system should permit the translation of input and output units into dollars to enable more accurate comparisons between factors of production, and with outside benchmarks.[mwf11]

Discussion[mwf12]

Method. Annual output and input data were obtained from Sawyer’s accounting department. Partial factor productivity (output/single input) was calculated for each factor of production. Percentage change in productivity year over year was then determined and compared to US manufacturing and nonfarm business data. We assumed[A13] that Sawyer’s operation is primarily involved in manufacturing, so the comparison to US manufacturing productivity data appears most relevant.[A14]

Productivity Results. [mwf15]Dallas plant productivity data appear in Table 1[mwf16]. Partial factor productivity, where output is compared to a single resource, was determined for each factor of production. Since little in-house historical productivity available for comparison, the most important data in Table 1 are likely the percentage changes in productivity between 1999 and 2000. Labor- and resin-related productivity increased significantly over the past year (9% and 11% [mwf17]respectively). A decrease in capital-related productivity (-9%) offset these gains somewhat.

It is often desirable to determine the productivity of multiple factors of production (multifactor productivity) or even all factors (total factor productivity). In order to do this for the Dallas plant, the accounting system should incorporate cost factors to convert units of production into their dollar equivalents. For instance[mwf18], the energy factor should be multiplied by the appropriate $/BTU cost factor to determine the dollar value of the resource. Note that converting input and output values into their dollar equivalents will also facilitate comparisons among various factors and with outside benchmarks.

Table 1: Dallas Billiard Plant Productivity[mwf19]
1999 / 2000
Output/Input Data
Units Produced / 1,000 / 1,000
Labor (hrs) / 300 / 275
Resin (lbs) / 50 / 45
Capital ($) / 10,000 / 11,000
Energy (BTU) / 3,000 / 2,850
Partial Factor Productivity / % Change
Labor (hrs) / 3.3 / 3.6 / 9%[mwf20]
Resin (lbs) / 20 / 22 / 11%
Capital ($) / 0.10 / 0.09 / -9%
Energy (BTU) / 0.33 / 0.35 / 5%
4%
note: partial factor productivity = output/single resource
source: Dallas plant accounting office[mwf21]

Outside Comparisons. The most widely followed national productivity data is the labor productivity series[mwf22] produced quarterly by the Bureau of Labor Statistics[mwf23]. Output from various sectors is compared against the number of corresponding labor hours. Note that this is a partial factor productivity, and unfortunately cannot be used to infer the productivity of other factors of production.

Figure 1 [mwf24]presents annual changes in US productivity for both the manufacturing and nonfarm business sectors. Note [mwf25]that, until recently, productivity has been increasing since about 1993. The Dallas plant’s labor productivity increase of 9% exceeds the 2-4% increases in manufacturing and nonfarm business productivity.

Note also from Figure 1 [mwf26]the consistent difference between manufacturing and nonfarm productivity changes. This has been a persistent national phenomenon and relates at least in part to the difficulty with measuring inputs and outputs in the service sector—which is represented in the nonfarm series. The Dallas plant productivity, of course, is more easily tied to the manufacturing sector.

Tracking System Development. Developing a system to track productivity of the Dallas plant on a quarterly basis is recommended[mwf27]. Such a system will provide a database for monitoring improvements in productivity as the facility implements new initiatives. It will also provide a warning signal to managers should productivity begin to falter. A system that combines both tabular data (similar to Table 1) with a picture of changes over time (like Figure 1) should offer a valuable management tool for understanding productivity levels and trends[mwf28].

Limitations. Dollar-denominated unit values for output, labor, resin, and energy were not available. Therefore, no determination of total factor productivity was done.

Sawyer Productivity AnalysisPage 1 of 3

[mwf1]1 Many memo heading formats are used in industry. This is just one of them. Microsoft Word includes a number of memo templates. Any of them are fine.

[mwf2]1 For all headings, I like to use just boldface—no caps or underlining. Boldface alone does a nice job of setting headings apart without being excessive.

[mwf3]1 This is usually the person(s) that requested the investigation. Could be more than one person. Note that, in addition to this header on the first page, all pages should include page numbers—usually best place at the bottom of the page.

[mwf4]1 This section should contain a couple of sentences to orient the reader. Background facts. Should tell the reader why you are writing the memo. Do NOT include a detailed table of contents or your conclusions. They are coming below.

[mwf5]1 Avoid the use of first person (“I conducted this study”) or second person (“You requested this study”). Stick to third person tense.

[mwf6]1 One note before leaving this section. Your role here is an analyst. Your statements should be cool and rational. Avoid editorial (“Productivity is critical to our business success”) or other slogans—your writers likely have their own opinions here and they probably don’t need yours. Maintain an unbiased, analytical stance.

[mwf7]1 The two or three key points you want to leave with your reader. No more than 3!! Readers who trust your abilities will likely go no further than this section.

[mwf8]1 My personal preference is to use paragraph headings in boldface to summarize key points in the Findings section. I’m also not much of a bullet point fan, although you may find them effective.

[mwf9]1 Manage how many numbers you put in this section. Most readers find it difficult to grasp a lot of numbers presented in paragraph form. Most of your numbers should appear below—likely in a table or graph.

[mwf10]1 Usually, one of your key points will be a recommendation or action item. Managers are decisions makers, and they usually sponsor an analysis like this one in order to make a decision or follow-up with some action.

[mwf11]1 NEVER let your findings sections spill over to page 2!

[mwf12]1 This section is for those people who need more detail than just your key points. Here, you present your analysis. Ideally, the reader should easily be able to link each key finding above with your detailed analysis.

[A13]In this section, also include any assumptions you are making in your analysis.

[A14]In this section, you want to tell the reader about your data and how you plan to analyze it. Include info on a) nature of data b) source c) method of analysis. If you are comparing alternatives (us/them, current/proposed, etc), state those alternative as well.

[mwf15]1 As noted above, I like to use paragraph headings to give readers a clue about the content of the upcoming section. Often, I make the labels in my Findings section and my Discussion section match to strengthen the link between the sections for the reader.

[mwf16]1 Numerical data are almost always best presented in a table. Note that I have referenced the table here.

[mwf17]1 Although the discussion section should be more detailed, I insert numerical data carefully into paragraph narrative. I only put key numbers I want to leave with the reader here—and leave the rest for the table.

[mwf18]1 Whenever you can provide an example to clarify what you’re talking about, do it. Amazingly, most managers have difficulty dealing with abstract concepts like the ones we learn in school. They learn much better by relating concepts to real life experiences or to concrete examples. Good analysis usually includes a number of sentences that begin with the terms “For example,…” or “For instance, …”

[mwf19]1 Admittedly, designing good tables is more of an art that usually comes thru experience. Some key points: always number and title, use space rather than heavy lines and colors to separate rows, columns, and sections. Also, don’t let your table get cut in half at a page break. Move it so that the entire table is on one page.

[mwf20]1 Even in calculated numbers, report no more than 3 significant digits. Most readers’ eyes glaze over when the digit string is long.

[mwf21]1 The bottom of a table is a great place to put equations, sample calculations, and sources of info.

[mwf22]1 Any data linked to time (years, months, days, etc) is usually called a ‘time series’ or just a ‘series.’ Series data is very common in business analysis.

[mwf23]1 Analysis is usually always enriched when the analyst goes ‘beyond’ the current problem for outside perspective. This is a lot easier now with the web, which is rich with government agencies, trade association data, and company websites.

[mwf24]1 Note again that I reference the graph in the narrative text. Some important characteristics of good graphs: titles and axis labels, a minimum of color, lines and other ‘gobbledygook’ that detracts from the message of the data. Unfortunately, by default, Excel creates graphs with lots of excessive ‘chartjunk’. Usually, by right-clicking graphs in Excel, you can easily edit default chart options. The secret of great graphs is to communicate your message (“Here are the relevant levels, trends, or comparisons”) without a lot of excessive ink (a.k.a. “chartjunk”).

[mwf25]1 Observe how nice it is to have the table or graph you reference close by. This is better than appending a bunch of attachments to the back of the memo (they rarely get read). The usability features of Microsoft Office allows an analyst to create tables and graphs in Excel, and then copy-and-paste them precisely into Word (I did this here).

[mwf26]1 One other note on tables and graphs: Table 1 and Figure 1 here were both developed in Excel, and then copied and pasted into the Word document.

[mwf27]1 As noted above, managers usually expect a recommendation or action item from analysis. Provide the reader with tips about how to implement the recommendation or its design. Sometimes, you might also volunteer to initiate it!

[mwf28]1 Most managers are constantly trying to grasp the level (averages) and trend (up or down) nature of their operations. Analyses like this one can help.