What sort of driving policy do you have? Are you a fast driver or a slow driver? And how should we evaluate a person’s driving policy?
Suppose I’m driving my Ferrari up a long mountain road outside of Denver. I overtake and fly by a slow moving 1968 VW minivan, which is chugging up the hill at 55 mph. I say to my passenger, “What a slow driver!” The passenger might respond by denying that I had passed a slow driver:
“Drivers should be judged based on intentions, not outcomes. Perhaps the minivan driver had depressed the gas pedal 80% of the way to the floor, while in this Ferrari the pedal is only depressed 50% of the way to the floor. In that case, the minivan driver had a faster speed policy, despite the slower outcome.”
That seems an odd way to describe a speed policy. But my passenger insists that the term “policy” implies actual intentions, not outcomes, and depressing the gas pedal by more or less is the specific action that causes the car to change speeds. So speed policy should be judged based on the specific actions taken by the driver.
I respond that depressing the gas pedal 80% of the way to the floor means something very different when going up a hill as compared to taking the same “concrete step” when going down a hill. It makes no sense to describe someone choosing to go 55 mph on an expressway as having a “fast speed policy.”
One potential compromise would be to define an equilibrium rate of pedal depression. That’s the gas pedal setting that yields the appropriate speed—say 70 mph. Using this criterion, depressing the gas pedal by more than equilibrium leads to a high speed policy, and depressing the gas pedal by less than equilibrium leads to a low speed policy.
I guess that would work, but how do you know the equilibrium rate of pedal depression? Don’t you have to look at the speedometer to determine whether the gas pedal has been depressed too much or too little? I’m not saying this pedal approach is impossible—maybe a computer could be programmed to take everything in account and estimate the pedal depression required to reach 70 mph, but isn’t it just easier to look at the speedometer?
In the comment section, please tell me if you would view a Ferrari and minivan as having the same speed policy if both drivers depressed the pedal by 50%, or would you regard these cars as having the same speed policy if both vehicles were going 70 mph?
Also, do you see any similarities between the question of ascertaining the speed policy of a driver, and ascertaining the monetary policy of a central bank? What is the analogy for pedal depression? What is the analogy for speedometer reading? What would the optimal speedometer look like if speed responds with a lag to pedal depression?
PS. Suppose a driver already going downhill depressed the gas pedal all the way to the floor. The car ended up going so fast that an accident resulted. Would it be a valid excuse to say, “The steep hill caused the accident”? How does this analogy apply to fiscal and monetary policy during 2021?
PPS. My actual car is not a Ferrari, it’s a Nissan Maxima. (Still faster than a minivan):
READER COMMENTS
Peter Gerdes
Dec 15 2023 at 7:16pm
This feels like a problem of definition and verbal shorthand.
Obviously, if we are being fully pedantic, a speed policy is a function from all relevant inputs to some output. Once you include enough inputs it really doesn’t matter if you choose to use speed pedal depression or speed as the output variable since a full specification of policy will include the fact that you’d depress the pedal to different levels depending on the car being driven.
Of course this doesn’t mean that in some contexts we might want to just know if the driver is attempting to accelerate or not. However, these all seem like concerns where the right move isn’t to argue over what is meant by policy in that context but just to ask for further clarification or to rephrase the question in terms of common terms.
Scott Sumner
Dec 15 2023 at 8:16pm
“This feels like a problem of definition”
Absolutely, it’s question of which definition is the most useful, and which definition is highly misleading.
Steve Fritzinger
Dec 15 2023 at 7:32pm
Setting your “policy” as either a specific speed or how far down you depress the accelerator is the Original Sin of highway driving. Your policy should be something like “maintain a 2 second distance between myself and the car in front of me”. How much you depress the accelerator is the instrument you use to implement your policy.
But you already knew that, right?
Scott Sumner
Dec 15 2023 at 8:17pm
Yes, although I think it’s reasonable to view speed as a policy on an uncrowded rural expressway.
Thomas L Hutcheson
Dec 15 2023 at 10:19pm
Even there the driver reacts to bend in the road, other vehicles. What is the need to discuss the driver’s “policy?” The driver is optimizing something subject to road and vehicle constraints. I guess we could call the speed that optimized the function at some arbitrary set of conditions the driver’s “policy.”
Scott Sumner
Dec 16 2023 at 12:41pm
Analogies are never perfect.
Kevin Corcoran
Dec 15 2023 at 8:55pm
What kind of driver am I? I’m an above average driver – just like everyone else!
Dick King
Dec 16 2023 at 11:54pm
It’s possible that the median driver has had no at fault accidents, but obviously the average driver has had more than zero. Therefore, the median driver is better than average.
-dk
Schepp
Dec 15 2023 at 8:59pm
The intent for speed is a function of the gas pedal, up or downhill and the capital cost of the Ferrari or VW.
The false smoothing of monetary signals hides the fact the check engine light should be on in both the Ferrari and the VW.
Thomas L Hutcheson
Dec 15 2023 at 10:11pm
How odd to discuss speed “policy.” Did the driver need to know the “policy” of the other driver in cedicing to overtake?
Re the PS:
The driver/Fed is supposed to use the accelerator/brake to keep the vehicle/economy going a the optimum trade off of speed and safety/ real income-maximizing inflation taking into account the changing road/economic conditions. We might define that optimum speed for a typical set of road conditions as FAST 🙂 ( Flexible Average Speed Target) of x mph analogous to the Fed’s FAIT.
Jim Glass
Dec 16 2023 at 2:43am
I’m thinking of rocketing a Cybertruck Elonmobile, 0-to-60 in 2.5 seconds and beyond, up along the winding cliff-edge of a narrow mountain road, rubber burning, hands free and waving them in the air thanks to the automatic Autopilot driving system, grokking pure joy …. AAAAHHHHHhhhhhhhh….
I don’t know. Is this a warning against relying the Taylor Rule or some other ‘automatic’ monetary guidance?
john hare
Dec 16 2023 at 4:03am
I am probably below average in some driving skills as I think my situational awareness is less capable than it was 30 years ago. As such, a bit of caution and common sense is useful. Run traffic average in the right lane of the interstate unless behind a rolling roadblock. Take an extra second to look and allow a bit more margin than I used to. Had one at fault highway accident in 52 years of driving including a lot of equipment trailers. Many minor scrapes at low speed on jobsites when jockeying around obstacles.
Bottom line is that a little patience, cooperation with others, and awareness of current limitations goes a long ways. No idea whatsoever how this would apply to monetary policy.
David S
Dec 16 2023 at 7:15am
My Toyota, like most modern vehicles, has a feature called “adaptive cruise control”–which works quite well in highway driving that isn’t too congested. It also does a better job of optimizing engine power output for efficient fuel consumption than I can do driving manually. However, the cruise control occasionally makes mistakes because it lacks a comprehensive understanding of the road environment. If it’s following a car that’s going slower than the desired speed then it adapts to the slower car. The radar also picks up cars that are exiting and slowing down and applies this bad information to slow my car as well.
In general, I can trust the cruise control about 85% of the time, but I need to stay vigilant as a driver and override the automation when it’s not performing as desired.
Dylan
Dec 16 2023 at 9:59am
I was in Costa Rica last year driving through some windy mountain passes on a nice (by Costa Rica standards) highway. There were several big box trucks that struggled mightily going up the hills, but as soon as they crested they would pick up speed dramatically, and pass me going around corners, leaning so dangerously to one side I thought they were going to tip. Only for me to be forced to pass them again once we were going up hill again.
I’d characterize those drivers as having a fast speed policy at all times, but external conditions, the slop of the hill and the vehicle type, prevented them from achieving it roughly half the time. I’ll leave it to others more attuned to monetary policy to see if this breaks the metaphor if we extend it to fed policy.
Scott Sumner
Dec 16 2023 at 12:44pm
In case anyone missed it, pedal depression is analogous to judging monetary policy by looking at interest rates, and the speedometer is analogous to NGDP growth. Ideally, you’d want a gauge that forecast how pedal depression would affect future speed—analogous to NGDP futures markets.
Comments are closed.