I don’t really see many emergencies that can be handled by cash but not by a loan for cash. If you’re solvent and people want dollars later, then they will lend you money. If you’re not solvent, then whether your immediate liquidity is in credit or cash doesn’t make a big difference since you’re still not solvent. If nobody wants dollars later (say, asteroid), then it’s unlikely that having dollars now is going to fix any emergencies.
If you’re solvent and people want dollars later, then they will lend you money.
I don’t find that obvious. There is a whole host of issues here, starting with time constraints (e.g. you need money within 24 hours and you can get a loan in five business days) and ending with information asymmetry issues of which lenders are acutely cognizant (“you say you’re solvent, but can you prove it?”).
If your “access to credit” is a couple of credit cards, yeah, you can get cash fast enough but the terms are rarely what I’d call “reasonable”. If you’d actually need a new loan or a line of credit… I don’t think I would want to rely on that in an emergency.
How often do you really need money within 24 hours? If you can’t get the cash within a day, what bad consequences are going to happen?
If it’s a purchase under $5000, then you can handle it with a credit card. You then have 21 days to come up with the money or else pay 20% APR. That’s plenty of time if you have, say, stocks you can sell. For larger purchases, you can either save for it with an explicit plan, or negotiate a payment plan.
How many times have you needed money immediately in your life, and how much money have you needed for those incidents? Personally, I do not recall ever spending more than a hundred dollars without at least a day’s warning. Then again, I don’t own a car, which is a big cause for emergency spending—but really that ought to have it’s own fund treated as self-insurance.
How many times have you needed money immediately in your life, and how much money have you needed for those incidents?
Well, if you want to approach this properly… :-)
...then you’ll need to evaluate the probability density of situations in your life where not having a certain amount of cash on hand will lead to severely negative outcomes (aka high costs). I expect that you’ll have much difficulty in trying to form a reasonable estimate (see Nassim Taleb and the general Black Swan concept). Notably, limited amount of historical data (as in, e.g. your personal experience) is not all that good a basis for estimations.
There is also a whole bunch of other factors in play—do you have kids? do you travel much? outside of the US? etc. etc.
Example 2: You live up north, it’s winter, and your house’s heating just died. If you don’t fix it by the time the house cools down to below freezing, some of your water pipes will burst.
That’s fair. I’d been thinking about the general class of “people who need money now for an emergency,” many of whom find it difficult to secure credit, rather than the class of “people who have a lot of wealth in non-liquid forms who need money now for an emergency,” who presumably don’t.
I was thinking more in terms of “there’s an expenses function e(t), and a cash availability function s(t), and a cost function f( e(t) - s(t) ), and this cost function is zero at e(t)-s(t) = 0, but is a lot softer at e(t) > s(t) than people fear due to credit cards and lines of credit, and can be quite costly at s(t) >> e(t)”
Except that e(t) and s(t) really should be probability distributions, but that just hurts my head to try and explain coherently. This is literally my fourth attempt at writing up a better description of the reasoning behind my posts.
If e(t) is slightly bigger than s(t), you borrow money from credit cards or other lines of credit at poor interest rates, then pay off those debts in the however many days it takes to get liquid cash from other sources (say, stocks). If e(t) is much bigger than s(t), then you negotiate a payment plan or suffer the consequences of not being able to pay expenses right now.
And of course there’s the time costs in optimizing this sort of thing. A percentage point for a thousand dollars over a year comes out to ten dollars, which I roughly approximate as an hour of time. Which means that you probably ought to spend your optimization power on minimizing the amount of work you need to put into your finances. Which, in turn, means automatic bill payment, and regular transfers of excess cash from your checking account into your preferred investment account.
I don’t really see many emergencies that can be handled by cash but not by a loan for cash. If you’re solvent and people want dollars later, then they will lend you money. If you’re not solvent, then whether your immediate liquidity is in credit or cash doesn’t make a big difference since you’re still not solvent. If nobody wants dollars later (say, asteroid), then it’s unlikely that having dollars now is going to fix any emergencies.
I don’t find that obvious. There is a whole host of issues here, starting with time constraints (e.g. you need money within 24 hours and you can get a loan in five business days) and ending with information asymmetry issues of which lenders are acutely cognizant (“you say you’re solvent, but can you prove it?”).
If your “access to credit” is a couple of credit cards, yeah, you can get cash fast enough but the terms are rarely what I’d call “reasonable”. If you’d actually need a new loan or a line of credit… I don’t think I would want to rely on that in an emergency.
How often do you really need money within 24 hours? If you can’t get the cash within a day, what bad consequences are going to happen?
If it’s a purchase under $5000, then you can handle it with a credit card. You then have 21 days to come up with the money or else pay 20% APR. That’s plenty of time if you have, say, stocks you can sell. For larger purchases, you can either save for it with an explicit plan, or negotiate a payment plan.
In an emergency I expect to need money right now, on the time scale of hours.
How many times have you needed money immediately in your life, and how much money have you needed for those incidents? Personally, I do not recall ever spending more than a hundred dollars without at least a day’s warning. Then again, I don’t own a car, which is a big cause for emergency spending—but really that ought to have it’s own fund treated as self-insurance.
Well, if you want to approach this properly… :-)
...then you’ll need to evaluate the probability density of situations in your life where not having a certain amount of cash on hand will lead to severely negative outcomes (aka high costs). I expect that you’ll have much difficulty in trying to form a reasonable estimate (see Nassim Taleb and the general Black Swan concept). Notably, limited amount of historical data (as in, e.g. your personal experience) is not all that good a basis for estimations.
There is also a whole bunch of other factors in play—do you have kids? do you travel much? outside of the US? etc. etc.
What sort of emergency do you have in mind?
Example 1: medevac.
Example 2: You live up north, it’s winter, and your house’s heating just died. If you don’t fix it by the time the house cools down to below freezing, some of your water pipes will burst.
Maybe you could drain all your pipes in the latter case. But I imagine there are other emergencies, of course.
That’s fair. I’d been thinking about the general class of “people who need money now for an emergency,” many of whom find it difficult to secure credit, rather than the class of “people who have a lot of wealth in non-liquid forms who need money now for an emergency,” who presumably don’t.
I was thinking more in terms of “there’s an expenses function e(t), and a cash availability function s(t), and a cost function f( e(t) - s(t) ), and this cost function is zero at e(t)-s(t) = 0, but is a lot softer at e(t) > s(t) than people fear due to credit cards and lines of credit, and can be quite costly at s(t) >> e(t)”
Except that e(t) and s(t) really should be probability distributions, but that just hurts my head to try and explain coherently. This is literally my fourth attempt at writing up a better description of the reasoning behind my posts.
If e(t) is slightly bigger than s(t), you borrow money from credit cards or other lines of credit at poor interest rates, then pay off those debts in the however many days it takes to get liquid cash from other sources (say, stocks). If e(t) is much bigger than s(t), then you negotiate a payment plan or suffer the consequences of not being able to pay expenses right now.
And of course there’s the time costs in optimizing this sort of thing. A percentage point for a thousand dollars over a year comes out to ten dollars, which I roughly approximate as an hour of time. Which means that you probably ought to spend your optimization power on minimizing the amount of work you need to put into your finances. Which, in turn, means automatic bill payment, and regular transfers of excess cash from your checking account into your preferred investment account.