My experience in firms from 20 - 150 employees:
1974-1984 - 1 engineer for every 2 to 4 drafters
1985-1995 - 1 engineer for every 1 to 2 drafters
1996-2007 - 3 engineers for every 1 drafter
2007 - present - engineers do the drafting

Currently all our engineers in the building design group do their own drawings - no drafters.

Benefits:
Direct control of your drawings and details
More intimate knowledge of what is on the plans
No need to QC back and forth between engineer and drafter
Less hours spent on projects
Less money spent on projects
Less time spent hiring/firing/laying off due to drafters that don't perform or quit.

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

Most of the drawing software platforms in widespread use today are really computer-aided engineering platforms, not just drafting platforms. For example, in one software, I can design control system schematics, select parts from a database library when I insert the component symbols, assign child contacts to parent coils and get a link created between them automatically, autonumber all components and wiring, lay out a control panel using those parts using their included shape files, generate reports like BOMs and error-checking reports, etc. I can also customize the symbols, databases, and shape files in a fairly easy fashion. In other words, I can do the engineering while I do the drafting, and I can do it in less time with fewer mistakes than in the old way of going back and forth with a drafter. It seems this is the way of the future.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips.

I think it depends partially on the type of work.

If your work is more repetitive (residential/commercial) or data intensive (BIM/BrIM modeling), you may find efficiencies in having a drafter at a lower salary. Assuming you can find a good one (getting harder).

The more unique the work, the more often I've seen engineers doing their own drafting. (for the reasons JAE stated)

----
The name is a long story -- just call me Lo.

The answer lies in whether you can find and do enough additional paying work to cover the additional cost. That depends on who you hire and what type of work you do (as others have said).

Are you turning down jobs now because you're too busy?

Quote (JAE)

2007 - present - engineers do the drafting

Count my firm as at least one data point against this. 30 person firm, engineers don't do much drafting. We have too much engineering work to do. If we had engineers do drafting we'd have to hire more engineers (at higher rates than drafters) or turn down work so engineers can spend time drafting. Ratio is about 4-5 engineers per drafter.

We have pretty low repetition work, and its almost impossible to efficiently use a drafter.

Doing my own drawings gives me "quality" (control) time with my designs.

Our firm, approximately 60 people, is a designers do design and drafters perform 100% of the drafting. For new design projects at least. For our restoration type work there are many of the designers that do their own drafting.

The ratio in the new design section is 2 designers to 1 drafter. Seems to keep everyone the right amount of swamped.

@BUGGAR
I agree with your thinking. I like to steep in my design as I draw it.

I've never had a draftsman to help my work. In many ways, I started my career as the drafter, before gaining trust to do the design/analysis/certification that builds upon it.
Now, I find I can visualize the concepts very well on my own, but details really drive much of my work, so I can't complete a design without CAD in some way to resolve how the details will affect the big picture. It always matters. I also do most stress analysis with the drawings open to get/check/adjust dimensions as I go. I really don't know how efficiently I would us a draftsman if I had one to help me.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

I beg to differ with some opinions out here about engineers doing their drafting.

1. When an engineer does his own drafting, he spends up to 70% of his time drafting and 30% on engineering.
2. you can QC yourself up to a point, there are more mistakes when an engineer does his own drafting than when done by drafter. Reason, you concentrate on commands, on lines, not on engineering, the result is that the engineering sucks. Big mistakes due to human errors occur. I know it about myself, I make mistakes that I do not do when someone else does the drafting. Look at your own drawing 2 days later and you find mistakes, not just drafting mistakes, engineering mistakes, because your mind was somewhere else.
3. There were some darn good drafters in the industry that caught quiet a few engineers mistakes, especially coordination with structural (HVAC world here)
4. After 10 years, an engineer doing his own drafting cannot possibly state that he has 10-years experience, he has only 4 years of engineering experience and 6 years of drafting experience i.e. 6-years of screen time.
5. Drafters are expert at CAD/revit, engineers only know enough. Work is done much faster. I think every MEP firm could use one or two drafter per discipline (say 2 Mech, 2 Elect, 1 Plumb) and come out on top for everyone, CAD guys makes a living and MEP firm gets work done.

And.. drafters are cheaper, much cheaper.

Same thing happens when you try to manage and design a project at the same time. Both management and engineering suffer.

CxA-Eng,
Some replies to your list:
1. Your percentage is off. I do my own drafting and this is not true at all.
2. I've worked both with drafters and with myself doing the drafting - there are more mistakes with drafting for the reason that there is always some level of communication breakdown between engineer and draftsman. And if you are using commands you need a software update.
3. Agree - there WERE some good "technicians" in the industry....but not as much anymore.
4. Disagree fully. The time spent sketching up details for draftsman, answering questions, following up with drawing QC, redlining a detail at least 3 times before complete all add up to MUCH MORE TIME than if you did it yourself.
5. My own engineers are "expert" in Revit and are very fast - as fast or faster than some 2 year drafting student from the local community college.

Drafters aren't cheaper. If you were a business owner you'd understand this - the cost of an employee is huge and we have found that engineers can draw on the computer and finish the design much faster than an engineer taking the time to communicate to a drafter and then follow up on the multiple checks.

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

JAE I agree with all your points.

It has been my experience that there are far more errors on drawings done by drafters. Only the very few, very best drafters are on par with what I would expect in terms of speed and low error count.

We have been trying to hire an experienced drafter for quite some time now, they just don’t exist anymore.

Currently, our engineers do all of their own drafting. I’m not a fan of this, as they have no formal CAD training, and they really aren’t all that good or efficient at it. I’d rather be paying one efficient drafter and have them focusing entirely on design, but like I said, drafters are starting to become unicorns.

Just before this degrades into a draftsman-bashing thread, a word in their defense:
Not many engineers get concerned with the clarity of presentation, consistency in a package of drawings, or orderliness in the views and cross-sections of an elaborate part or installation. For adherence to standards, and correction of confusing presentation, I can't do better than to have an experienced draftsman to check my drawings.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

I'm certainly not intending to "bash" draftsmen. I've worked with many who were very good.

But with the cadd/Revit of today, and the much better interface in the software, they are not as essential, or efficient, as they once were.

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

I think part of the argument about efficiency is simply that these days drafter are taught to simply use the software in formal education, and are not taught anything about structures or how to draw them, how they work, etc like they were back in the day. Even when they get on the job as a graduate they are not taught these aspects. If you want them to be good you have to give them an environment to get good.

Back in the day its was the opposite, older types I've worked with for example tell stories about how the first 3 month on the job were spent printing text again and again in the drawing office until it was perfect and they were ready to be unleashed on a real world job. This care and attention are something that seems to have been lost along the way.

I've worked with 'drafters' and 'tracers', each have their uses, but I'd much prefer to draw it myself or utilise the most experienced drafters (those guys who are worth the money they are paid). If using less experienced people and grumbling about the quality/time/efficiency, then take a moment to teach them why they are drawing it the way you want to see it and they will understand for next time, or alternatively listen to the way they want to draw it and you might learn something.

I would never want to be without good drafting and checking departments simply bc I've got more than enough higher level engineering tasks to do. Some months I release a couple dozen prints and others I release several hundred, so having a team of specialists able to jump on those tasks is a serious timesaver and keeps our engineers focused on legitimate engineering, not worrying over bolted joints, tolerance stacks, and other minor details. Having received some pretty lousy supplier prints over the years, I also believe that proper draftsmen do a much better job and ultimately print quality is critical to business.

CWB1 - this all might depend a lot on the nature, discipline, and size, of the projects.
So in your case it may be good to have "draftsmen/women" checking your designs, documents, etc.
For us it would be an inefficient process.

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

Quote (JAE)

This all might depend a lot on the nature, discipline, and size, of the projects.

One metric for that is the ratio of drawings you release to an outside customer to drawings released internally within the company.
The value of draftsmanship has been directly proportional to the rate of drawings released to outside customers, IME.
I have worked at a company that released almost 100% of its drawings externally, another 50/50 internal/external, and a third whose drawings were used 100% internally.
The drafting emphasis changed drastically with the change of "customer".

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

JAE, I concur. Simply wanted to add a data point and mention that our use of draftsmen isn't limited to simply creating the print. An experienced draftsmen can often be a sanity check for junior engineers and handle many of the same tasks.

I work in a very small firm that handles a wide range of projects form small residential up to multi-story commercial/office.

We are all engineers and we all do our own drafting. Some are better than others. We just recently started transitioning over to Revit from using only Autocad, but the transition has brought us more work. I think the main benefit of the engineer doing both the design and the drafting is that you become intimately familiar with how the details go together. This helps with not producing "unbuildable" drawings or drawings that work fine on paper. The understanding that generating your own drawings creates IMHO leads to higher quality drawings. As a side note, we have always done our own drafting and therefore, we are very proficient. Maybe not as a proficient as an experienced drafter, but I think the pros outweight the cons.

StrucDesignEIT just touched on a good point - the recent surge in BIM designs for buildings has significantly changed how designs are created "on paper".

Autocad is generally a bunch of dumb lines where you can tag some data.
Revit is a database that happens to be able to draw, or model in 3D, its subjects.

In using Revit, the interface between the design, the calculations, the representation of the building, the materials, etc. all come into play very early in the design process.
This is not very adaptable for simple drafting personnel. The "drawings" develop as the design develops. There are interfaces between the Revit model and the analysis software.
None of this works with a drafting person - at least not very well.

Once the mode is completed, I suppose a drafter could then start putting together plans, sections and details, but the engineer is so familiar with it by then that it would be foolish to then turn the model over to a drafter with no understanding of the deaign, the data, and the model.

Check out Eng-Tips Forum's Policies here:
FAQ731-376: Eng-Tips.com Forum Policies

I am a solo operator (structural - mostly heavy industrial) and pretty much every project is unique. I have tried several times to work with drafters but usually find it takes almost as much time for me to explain what to do and then run through several rounds of corrections. The biggest issue may be my end because I design/draft/and engineer at the same time.

Most recently I have found a very experienced and knowledgeable drafter. I have done a few small projects where I start the drawing and then have him finish out by adding sections, detail views and clean up the presentation. While he has kept his end of the work on my side I often make small changes to beam sizes or connection details as I see how the final design is coming together.

I have had more success with doing larger projects in Tekla. I build the 3D model and do the GA's so my client can get bids. I then share the model with some outside detailers and they clean up the model and do the fabrication drawings (E, Assembly and single part). Tekla's model sharing allows me to monitor their progress and make changes as needed.

I learned Revit and would love to use BIM on a project but the complexity (and costs) involved in coordinating with the other trades, product version incompatibility, learning curve, subscription evils, etc., I subconsult my BIM work, when required, to others who specialize in it. BIM is supposed to benefit the owner and why shouldn't it be a separate service that "lives" on after the engineering is complete? This is what the BIM salesmen say.

I like having a drafter "clean" up my design drawings and do the reality check on them.

It's a division of labour thing. If you're doing computer-aided design, having the engineer do the work is essential. If you're preparing 2D drawings, a drafter can do the work for less money for your client.

As to whether or not you can afford to have a drafter in a 3 person office- I'd imagine that a sub would be a far better option for you at that size because unless that drafter can do other work, they're going to be twiddling their thumbs a lot.